US20210343286A1 - Auto-completion for Multi-modal User Input in Assistant Systems - Google Patents

Abstract

In one embodiment, a method includes receiving an initial input in a first modality from a first user at a client system, determining intents and slots corresponding to the initial input, wherein the slots are conditioned on the intents, generating one or more candidate continuation-inputs based on the intents and slots, where the one or more candidate continuation-inputs are in one or more candidate modalities, respectively, wherein the candidate modalities are different from the first modality, and wherein each of the candidate continuation-inputs references entities represented by the slots, and presenting one or more suggested inputs corresponding to one or more of the candidate continuation-inputs at the client system.

Description

PRIORITY
• This application is a continuation under 35 U.S.C. § 120 of U.S. patent application Ser. No. 16/389,728, filed 19 Apr. 2019, which claims the benefit, under 35 U.S.C. § 119(e), of U.S. Provisional Patent Application No. 62/660,876, filed 20 Apr. 2018, each of which is incorporated herein by reference.
TECHNICAL FIELD
• This disclosure generally relates to dialog management based on machine-learning techniques within network environments, and in particular relates to hardware and software for smart assistant systems.
BACKGROUND
• An assistant system can provide information or services on behalf of a user based on a combination of user input, location awareness, and the ability to access information from a variety of online sources (such as weather conditions, traffic congestion, news, stock prices, user schedules, retail prices, etc.). The user input may include text (e.g., online chat), especially in an instant messaging application or other applications, voice, images, motion, or a combination of them. The assistant system may perform concierge-type services (e.g., making dinner reservations, purchasing event tickets, making travel arrangements) or provide information based on the user input. The assistant system may also perform management or data-handling tasks based on online information and events without user initiation or interaction. Examples of those tasks that may be performed by an assistant system may include schedule management (e.g., sending an alert to a dinner date that a user is running late due to traffic conditions, update schedules for both parties, and change the restaurant reservation time). The assistant system may be enabled by the combination of computing devices, application programming interfaces (APIs), and the proliferation of applications on user devices.
• A social-networking system, which may include a social-networking website, may enable its users (such as persons or organizations) to interact with it and with each other through it. The social-networking system may, with input from a user, create and store in the social-networking system a user profile associated with the user. The user profile may include demographic information, communication-channel information, and information on personal interests of the user. The social-networking system may also, with input from a user, create and store a record of relationships of the user with other users of the social-networking system, as well as provide services (e.g. profile/news feed posts, photo-sharing, event organization, messaging, games, or advertisements) to facilitate social interaction between or among users.
• The social-networking system may send over one or more networks content or messages related to its services to a mobile or other computing device of a user. A user may also install software applications on a mobile or other computing device of the user for accessing a user profile of the user and other data within the social-networking system. The social-networking system may generate a personalized set of content objects to display to a user, such as a newsfeed of aggregated stories of other users connected to the user.
SUMMARY OF PARTICULAR EMBODIMENTS
• In particular embodiments, the assistant system may assist a user to obtain information or services. The assistant system may enable the user to interact with it with multi-modal user input (such as voice, text, image, video, motion) in stateful and multi-turn conversations to get assistance. The assistant system may create and store a user profile comprising both personal and contextual information associated with the user. In particular embodiments, the assistant system may analyze the user input using natural-language understanding. The analysis may be based on the user profile for more personalized and context-aware understanding. The assistant system may resolve entities associated with the user input based on the analysis. In particular embodiments, the assistant system may interact with different agents to obtain information or services that are associated with the resolved entities. The assistant system may generate a response for the user regarding the information or services by using natural-language generation. Through the interaction with the user, the assistant system may use dialog management techniques to manage and forward the conversation flow with the user. In particular embodiments, the assistant system may further assist the user to effectively and efficiently digest the obtained information by summarizing the information. The assistant system may also assist the user to be more engaging with an online social network by providing tools that help the user interact with the online social network (e.g., creating posts, comments, messages). The assistant system may additionally assist the user to manage different tasks such as keeping track of events. In particular embodiments, the assistant system may proactively execute tasks that are relevant to user interests and preferences based on the user profile without a user input. In particular embodiments, the assistant system may check privacy settings to ensure that accessing a user's profile or other user information and executing different tasks are permitted subject to the user's privacy settings.
• In particular embodiments, the assistant system may receive an initial input in a first modality executed by a user as an interaction with an assistant-based client system, e.g., a gesture-input to a virtual reality (VR) headset or augment reality (AR) smart glasses, determine candidate continuation-inputs in an auto-completion manner based on the initial input, and recommend these candidate continuation-inputs to the user in one or more second modalities (e.g., text), for which the user may select one of them to trigger the execution of a particular task. The initial input and continuation-inputs may be based on any suitable modality including text, speech, image, video, motion, orientation, etc. In addition, the modality of the initial input and the modalities of the continuation-inputs may be different. As an example and not by way of limitation, a user visiting Paris who wears AR glasses may be looking at Eiffel Tower. Based on the gaze input, the assistant system may suggest a gesture displayed on the virtual screen for taking a picture or suggest the user to speak “take a picture.” The user may perform the suggested gesture or speak the sentence to allow the assistant system to execute the task of taking a picture of Eiffel Tower. As a result, the assistant system may have the ability to process a user input in one modality and generate suggested inputs in one or more other modalities, which may be referred as auto-completion for multi-modal user input. Although this disclosure describes generating particular auto-completion for particular multi-modal user inputs via particular systems in particular manners, this disclosure contemplates generating any suitable auto-completion for any suitable multi-modal user input via any suitable system in any suitable manner.
• In particular embodiments, the assistant system may receive, from a client system associated with a first user, an initial input from the first user. The initial input may be in a first modality. In particular embodiments, the assistant system may determine, by an intent-understanding module, one or more intents corresponding to the initial input. The assistant system may then generate, based on the one or more intents, one or more candidate continuation-inputs. The one or more candidate continuation-inputs may be in one or more candidate modalities, respectively. The candidate modalities may be different from the first modality. In particular embodiments, the assistant system may further send, to the client system, instructions for presenting one or more suggested inputs corresponding to one or more of the candidate continuation-inputs.
• In particular embodiments, the assistant system may receive an incomplete gesture executed by a user as an interaction with an assistant-based client system, e.g., a VR headset or AR smart glasses, determine candidate gestures in an auto-completion manner based on the incomplete gesture, recommend these candidate gestures to the user, for which the user may select one of them to trigger the execution of a particular task. As an example and not by way of limitation, a user wearing AR glasses may not know what gesture to perform to trigger a particular function of the assistant system. The user may start moving his/her hand but pause in the air. Accordingly, the assistant system may analyze the incomplete gesture of the user and determine possible candidate gestures. The assistant system may display these candidate gestures visually via the AR glasses to the user. After the user selects one of them, the assistant system may further execute a task corresponding to that gesture. In summary, the assistant system may use gesture recognition techniques on incomplete gestures to determine the user-intended gestures and suggest the determined gestures as guidance to the user, which may be referred as auto-completion for gesture-input. Although this disclosure describes generating particular auto-completion for particular gesture-inputs via particular systems in particular manners, this disclosure contemplates generating any suitable auto-completion for any suitable gesture-input via any suitable system in any suitable manner.
• In particular embodiments, the assistant system may receive, from a client system associated with a first user, a user input from the first user. The user input may comprise an incomplete gesture performed by the first user. In particular embodiments, the assistant system may calculate, by an intent-understanding module, one or more confidences scores for one or more intents corresponding to the incomplete gesture. The assistant system may then determine that the calculated confidence scores associated with each of the intents are below a threshold score. In particular embodiments, the assistant system may select, based on a personalized gesture-recognition model, one or more candidate gestures from a plurality of pre-defined gestures responsive to determining that the calculated confidence scores for each of the intents are below the threshold score. Each of the candidate gestures may be associated with a confidence score representing a likelihood the first user intended to input the respective candidate gesture. In particular embodiments, the assistant system may further send, to the client system, instructions for presenting one or more suggested inputs corresponding to one or more of the candidate gestures.
• Certain technical challenges exist for achieving the goal of suggesting multi-modal user input for auto-completion. One technical challenge may include determining candidate continuation-inputs based on the initial input. A solution presented by the embodiments disclosed herein to address the above challenge is determining the candidate continuation-inputs based on the intent of the initial input and the potential entities relevant to the intent, or an object associated with the initial input, as such information may be necessary to determine what continuation-input a user may use to interact with the assistant system to execute tasks that based on the intent and the entities or the object. Another technical challenge may include determining that a user needs suggested inputs. A solution presented by the embodiments disclosed herein to address this challenge is determining whether to suggest candidate continuation-inputs based on different factors including wake-up inputs, gaze information, contextual information, and intents, as such information may reveal a user's requirement for assistance from different perspectives.
• Certain embodiments disclosed herein may provide one or more technical advantages. A technical advantage of the embodiments may include enriching user experience with the assistant system as a user may interact with the assistant system with inputs in various modalities. Another technical advantage of the embodiments may include increasing the degree of users engaging with the assistant system by guiding users with sequences of suggested inputs to explore various tasks that the users may request assistance. Certain embodiments disclosed herein may provide none, some, or all of the above technical advantages. One or more other technical advantages may be readily apparent to one skilled in the art in view of the figures, descriptions, and claims of the present disclosure.
• Certain technical challenges exist for achieving the goal of suggesting candidate gestures for auto-completion. One technical challenge includes determining similarity levels of candidate gestures with respect to the incomplete gesture. The solution presented by the embodiments disclosed herein to address the above challenge is determining the similarity levels based on different factors including trajectory, orientation, objects, contextual information, and position associated with the incomplete gesture, as such information may provide different informative cues for the assistant system to calculate similarity levels. Another technical challenge includes determining that a user needs suggested inputs for gestures. The solution presented by the embodiments disclosed herein to address this challenge is determining whether to suggest candidate gestures based on different factors including confidence scores of intents, wake-up gestures, temporal information of the incomplete gesture, velocity of the incomplete gesture, and patterns of the incomplete gesture, as such information may reveal a user's requirement for assistance from different perspectives.
• Certain embodiments disclosed herein may provide one or more technical advantages. A technical advantage of the embodiments may include improving user experience with the assistant system as the assistant system may understand a user's intent based on an incomplete gesture and teach a user to complete the incomplete gesture to interact with the assistant system. Another technical advantage of the embodiments may include increasing the degree of users engaging with the assistant system by guiding users with sequences of suggested gestures to explore various tasks that the users may request assistance. Certain embodiments disclosed herein may provide none, some, or all of the above technical advantages. One or more other technical advantages may be readily apparent to one skilled in the art in view of the figures, descriptions, and claims of the present disclosure.
• The embodiments disclosed herein are only examples, and the scope of this disclosure is not limited to them. Particular embodiments may include all, some, or none of the components, elements, features, functions, operations, or steps of the embodiments disclosed herein. Embodiments according to the invention are in particular disclosed in the attached claims directed to a method, a storage medium, a system and a computer program product, wherein any feature mentioned in one claim category, e.g. method, can be claimed in another claim category, e.g. system, as well. The dependencies or references back in the attached claims are chosen for formal reasons only. However any subject matter resulting from a deliberate reference back to any previous claims (in particular multiple dependencies) can be claimed as well, so that any combination of claims and the features thereof are disclosed and can be claimed regardless of the dependencies chosen in the attached claims. The subject-matter which can be claimed comprises not only the combinations of features as set out in the attached claims but also any other combination of features in the claims, wherein each feature mentioned in the claims can be combined with any other feature or combination of other features in the claims. Furthermore, any of the embodiments and features described or depicted herein can be claimed in a separate claim and/or in any combination with any embodiment or feature described or depicted herein or with any of the features of the attached claims.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 illustrates an example network environment associated with an assistant system.
• FIG. 2 illustrates an example architecture of the assistant system.
• FIG. 3 illustrates an example diagram flow of responding to a user request by the assistant system.
• FIG. 4 illustrates an example workflow of processing a user input comprising a gesture-input.
• FIG. 5 illustrates an example scenario of auto-completion for multi-modal user input.
• FIG. 6 illustrates an example scenario of auto-completion for gesture-input.
• FIG. 7A illustrates an example scenario of an incomplete gesture in AR/VR setting.
• FIG. 7B illustrates an example scenario of suggested gesture-input in AR/VR setting.
• FIG. 8 illustrates an example method for suggesting multi-modal user input for auto-completion.
• FIG. 9 illustrates an example method for suggesting candidate gestures for auto-completion.
• FIG. 10 illustrates an example social graph.
• FIG. 11 illustrates an example view of an embedding space.
• FIG. 12 illustrates an example artificial neural network.
• FIG. 13 illustrates an example computer system.
DESCRIPTION OF EXAMPLE EMBODIMENTSSystem Overview
• FIG. 1 illustrates anexample network environment100 associated with an assistant system.Network environment100 includes aclient system130, anassistant system140, a social-networking system160, and a third-party system170 connected to each other by anetwork110. AlthoughFIG. 1 illustrates a particular arrangement of aclient system130, anassistant system140, a social-networking system160, a third-party system170, and anetwork110, this disclosure contemplates any suitable arrangement of aclient system130, anassistant system140, a social-networking system160, a third-party system170, and anetwork110. As an example and not by way of limitation, two or more of aclient system130, a social-networking system160, anassistant system140, and a third-party system170 may be connected to each other directly, bypassing anetwork110. As another example, two or more of aclient system130, anassistant system140, a social-networking system160, and a third-party system170 may be physically or logically co-located with each other in whole or in part. Moreover, althoughFIG. 1 illustrates a particular number ofclient systems130,assistant systems140, social-networking systems160, third-party systems170, andnetworks110, this disclosure contemplates any suitable number ofclient systems130,assistant systems140, social-networking systems160, third-party systems170, and networks110. As an example and not by way of limitation,network environment100 may includemultiple client systems130,assistant systems140, social-networking systems160, third-party systems170, and networks110.
• This disclosure contemplates anysuitable network110. As an example and not by way of limitation, one or more portions of anetwork110 may include an ad hoc network, an intranet, an extranet, a virtual private network (VPN), a local area network (LAN), a wireless LAN (WLAN), a wide area network (WAN), a wireless WAN (WWAN), a metropolitan area network (MAN), a portion of the Internet, a portion of the Public Switched Telephone Network (PSTN), a cellular telephone network, or a combination of two or more of these. Anetwork110 may include one ormore networks110.
• Links150 may connect aclient system130, anassistant system140, a social-networking system160, and a third-party system170 to acommunication network110 or to each other. This disclosure contemplates anysuitable links150. In particular embodiments, one ormore links150 include one or more wireline (such as for example Digital Subscriber Line (DSL) or Data Over Cable Service Interface Specification (DOCSIS)), wireless (such as for example Wi-Fi or Worldwide Interoperability for Microwave Access (WiMAX)), or optical (such as for example Synchronous Optical Network (SONET) or Synchronous Digital Hierarchy (SDH)) links. In particular embodiments, one ormore links150 each include an ad hoc network, an intranet, an extranet, a VPN, a LAN, a WLAN, a WAN, a WWAN, a MAN, a portion of the Internet, a portion of the PSTN, a cellular technology-based network, a satellite communications technology-based network, anotherlink150, or a combination of two or moresuch links150.Links150 need not necessarily be the same throughout anetwork environment100. One or morefirst links150 may differ in one or more respects from one or moresecond links150.
• In particular embodiments, aclient system130 may be an electronic device including hardware, software, or embedded logic components or a combination of two or more such components and capable of carrying out the appropriate functionalities implemented or supported by aclient system130. As an example and not by way of limitation, aclient system130 may include a computer system such as a desktop computer, notebook or laptop computer, netbook, a tablet computer, e-book reader, GPS device, camera, personal digital assistant (PDA), handheld electronic device, cellular telephone, smartphone, smart speaker, virtual reality (VR) headset, augment reality (AR) smart glasses, other suitable electronic device, or any suitable combination thereof. In particular embodiments, theclient system130 may be a smart assistant device. More information on smart assistant devices may be found in U.S. patent application Ser. No. 15/949,011, filed 9 Apr. 2018, U.S. Patent Application No. 62/655,751, filed 10 Apr. 2018, U.S. Design patent application No. 29/631,910, filed 3 Jan. 2018, U.S. Design patent application No. 29/631,747, filed 2 Jan. 2018, U.S. Design patent application No. 29/631,913, filed 3 Jan. 2018, and U.S. Design patent application No. 29/631,914, filed 3 Jan. 2018, each of which is incorporated by reference. This disclosure contemplates anysuitable client systems130. Aclient system130 may enable a network user at aclient system130 to access anetwork110. Aclient system130 may enable its user to communicate with other users atother client systems130.
• In particular embodiments, aclient system130 may include aweb browser132, and may have one or more add-ons, plug-ins, or other extensions. A user at aclient system130 may enter a Uniform Resource Locator (URL) or other address directing aweb browser132 to a particular server (such asserver162, or a server associated with a third-party system170), and theweb browser132 may generate a Hyper Text Transfer Protocol (HTTP) request and communicate the HTTP request to server. The server may accept the HTTP request and communicate to aclient system130 one or more Hyper Text Markup Language (HTML) files responsive to the HTTP request. Theclient system130 may render a web interface (e.g. a webpage) based on the HTML files from the server for presentation to the user. This disclosure contemplates any suitable source files. As an example and not by way of limitation, a web interface may be rendered from HTML files, Extensible Hyper Text Markup Language (XHTML) files, or Extensible Markup Language (XML) files, according to particular needs. Such interfaces may also execute scripts, combinations of markup language and scripts, and the like. Herein, reference to a web interface encompasses one or more corresponding source files (which a browser may use to render the web interface) and vice versa, where appropriate.
• In particular embodiments, aclient system130 may include a social-networking application134 installed on theclient system130. A user at aclient system130 may use the social-networking application134 to access on online social network. The user at theclient system130 may use the social-networking application134 to communicate with the user's social connections (e.g., friends, followers, followed accounts, contacts, etc.). The user at theclient system130 may also use the social-networking application134 to interact with a plurality of content objects (e.g., posts, news articles, ephemeral content, etc.) on the online social network. As an example and not by way of limitation, the user may browse trending topics and breaking news using the social-networking application134.
• In particular embodiments, aclient system130 may include anassistant application136. A user at aclient system130 may use theassistant application136 to interact with theassistant system140. In particular embodiments, theassistant application136 may comprise a stand-alone application. In particular embodiments, theassistant application136 may be integrated into the social-networking application134 or another suitable application (e.g., a messaging application). In particular embodiments, theassistant application136 may be also integrated into theclient system130, an assistant hardware device, or any other suitable hardware devices. In particular embodiments, theassistant application136 may be accessed via theweb browser132. In particular embodiments, the user may provide input via different modalities. As an example and not by way of limitation, the modalities may include audio, text, image, video, motion, orientation, etc. Theassistant application136 may communicate the user input to theassistant system140. Based on the user input, theassistant system140 may generate responses. Theassistant system140 may send the generated responses to theassistant application136. Theassistant application136 may then present the responses to the user at theclient system130. The presented responses may be based on different modalities such as audio, text, image, and video. As an example and not by way of limitation, the user may verbally ask theassistant application136 about the traffic information (i.e., via an audio modality). Theassistant application136 may then communicate the request to theassistant system140. Theassistant system140 may accordingly generate the result and send it back to theassistant application136. Theassistant application136 may further present the result to the user in text.
• In particular embodiments, anassistant system140 may assist users to retrieve information from different sources. Theassistant system140 may also assist user to request services from different service providers. In particular embodiments, theassist system140 may receive a user request for information or services via theassistant application136 in theclient system130. Theassist system140 may use natural-language understanding to analyze the user request based on user's profile and other relevant information. The result of the analysis may comprise different entities associated with an online social network. Theassistant system140 may then retrieve information or request services associated with these entities. In particular embodiments, theassistant system140 may interact with the social-networking system160 and/or third-party system170 when retrieving information or requesting services for the user. In particular embodiments, theassistant system140 may generate a personalized communication content for the user using natural-language generating techniques. The personalized communication content may comprise, for example, the retrieved information or the status of the requested services. In particular embodiments, theassistant system140 may enable the user to interact with it regarding the information or services in a stateful and multi-turn conversation by using dialog-management techniques. The functionality of theassistant system140 is described in more detail in the discussion ofFIG. 2 below.
• In particular embodiments, the social-networking system160 may be a network-addressable computing system that can host an online social network. The social-networking system160 may generate, store, receive, and send social-networking data, such as, for example, user-profile data, concept-profile data, social-graph information, or other suitable data related to the online social network. The social-networking system160 may be accessed by the other components ofnetwork environment100 either directly or via anetwork110. As an example and not by way of limitation, aclient system130 may access the social-networking system160 using aweb browser132, or a native application associated with the social-networking system160 (e.g., a mobile social-networking application, a messaging application, another suitable application, or any combination thereof) either directly or via anetwork110. In particular embodiments, the social-networking system160 may include one ormore servers162. Eachserver162 may be a unitary server or a distributed server spanning multiple computers or multiple datacenters.Servers162 may be of various types, such as, for example and without limitation, web server, news server, mail server, message server, advertising server, file server, application server, exchange server, database server, proxy server, another server suitable for performing functions or processes described herein, or any combination thereof. In particular embodiments, eachserver162 may include hardware, software, or embedded logic components or a combination of two or more such components for carrying out the appropriate functionalities implemented or supported byserver162. In particular embodiments, the social-networking system160 may include one ormore data stores164.Data stores164 may be used to store various types of information. In particular embodiments, the information stored indata stores164 may be organized according to specific data structures. In particular embodiments, eachdata store164 may be a relational, columnar, correlation, or other suitable database. Although this disclosure describes or illustrates particular types of databases, this disclosure contemplates any suitable types of databases. Particular embodiments may provide interfaces that enable aclient system130, a social-networking system160, anassistant system140, or a third-party system170 to manage, retrieve, modify, add, or delete, the information stored indata store164.
• In particular embodiments, the social-networking system160 may store one or more social graphs in one ormore data stores164. In particular embodiments, a social graph may include multiple nodes—which may include multiple user nodes (each corresponding to a particular user) or multiple concept nodes (each corresponding to a particular concept)—and multiple edges connecting the nodes. The social-networking system160 may provide users of the online social network the ability to communicate and interact with other users. In particular embodiments, users may join the online social network via the social-networking system160 and then add connections (e.g., relationships) to a number of other users of the social-networking system160 whom they want to be connected to. Herein, the term “friend” may refer to any other user of the social-networking system160 with whom a user has formed a connection, association, or relationship via the social-networking system160.
• In particular embodiments, the social-networking system160 may provide users with the ability to take actions on various types of items or objects, supported by the social-networking system160. As an example and not by way of limitation, the items and objects may include groups or social networks to which users of the social-networking system160 may belong, events or calendar entries in which a user might be interested, computer-based applications that a user may use, transactions that allow users to buy or sell items via the service, interactions with advertisements that a user may perform, or other suitable items or objects. A user may interact with anything that is capable of being represented in the social-networking system160 or by an external system of a third-party system170, which is separate from the social-networking system160 and coupled to the social-networking system160 via anetwork110.
• In particular embodiments, the social-networking system160 may be capable of linking a variety of entities. As an example and not by way of limitation, the social-networking system160 may enable users to interact with each other as well as receive content from third-party systems170 or other entities, or to allow users to interact with these entities through an application programming interfaces (API) or other communication channels.
• In particular embodiments, a third-party system170 may include one or more types of servers, one or more data stores, one or more interfaces, including but not limited to APIs, one or more web services, one or more content sources, one or more networks, or any other suitable components, e.g., that servers may communicate with. A third-party system170 may be operated by a different entity from an entity operating the social-networking system160. In particular embodiments, however, the social-networking system160 and third-party systems170 may operate in conjunction with each other to provide social-networking services to users of the social-networking system160 or third-party systems170. In this sense, the social-networking system160 may provide a platform, or backbone, which other systems, such as third-party systems170, may use to provide social-networking services and functionality to users across the Internet.
• In particular embodiments, a third-party system170 may include a third-party content object provider. A third-party content object provider may include one or more sources of content objects, which may be communicated to aclient system130. As an example and not by way of limitation, content objects may include information regarding things or activities of interest to the user, such as, for example, movie show times, movie reviews, restaurant reviews, restaurant menus, product information and reviews, or other suitable information. As another example and not by way of limitation, content objects may include incentive content objects, such as coupons, discount tickets, gift certificates, or other suitable incentive objects.
• In particular embodiments, the social-networking system160 also includes user-generated content objects, which may enhance a user's interactions with the social-networking system160. User-generated content may include anything a user can add, upload, send, or “post” to the social-networking system160. As an example and not by way of limitation, a user communicates posts to the social-networking system160 from aclient system130. Posts may include data such as status updates or other textual data, location information, photos, videos, links, music or other similar data or media. Content may also be added to the social-networking system160 by a third-party through a “communication channel,” such as a newsfeed or stream.
• In particular embodiments, the social-networking system160 may include a variety of servers, sub-systems, programs, modules, logs, and data stores. In particular embodiments, the social-networking system160 may include one or more of the following: a web server, action logger, API-request server, relevance-and-ranking engine, content-object classifier, notification controller, action log, third-party-content-object-exposure log, inference module, authorization/privacy server, search module, advertisement-targeting module, user-interface module, user-profile store, connection store, third-party content store, or location store. The social-networking system160 may also include suitable components such as network interfaces, security mechanisms, load balancers, failover servers, management-and-network-operations consoles, other suitable components, or any suitable combination thereof. In particular embodiments, the social-networking system160 may include one or more user-profile stores for storing user profiles. A user profile may include, for example, biographic information, demographic information, behavioral information, social information, or other types of descriptive information, such as work experience, educational history, hobbies or preferences, interests, affinities, or location. Interest information may include interests related to one or more categories. Categories may be general or specific. As an example and not by way of limitation, if a user “likes” an article about a brand of shoes the category may be the brand, or the general category of “shoes” or “clothing.” A connection store may be used for storing connection information about users. The connection information may indicate users who have similar or common work experience, group memberships, hobbies, educational history, or are in any way related or share common attributes. The connection information may also include user-defined connections between different users and content (both internal and external). A web server may be used for linking the social-networking system160 to one ormore client systems130 or one or more third-party systems170 via anetwork110. The web server may include a mail server or other messaging functionality for receiving and routing messages between the social-networking system160 and one ormore client systems130. An API-request server may allow anassistant system140 and a third-party system170 to access information from the social-networking system160 by calling one or more APIs. An action logger may be used to receive communications from a web server about a user's actions on or off the social-networking system160. In conjunction with the action log, a third-party-content-object log may be maintained of user exposures to third-party-content objects. A notification controller may provide information regarding content objects to aclient system130. Information may be pushed to aclient system130 as notifications, or information may be pulled from aclient system130 responsive to a request received from aclient system130. Authorization servers may be used to enforce one or more privacy settings of the users of the social-networking system160. A privacy setting of a user determines how particular information associated with a user can be shared. The authorization server may allow users to opt in to or opt out of having their actions logged by the social-networking system160 or shared with other systems (e.g., a third-party system170), such as, for example, by setting appropriate privacy settings. Third-party-content-object stores may be used to store content objects received from third parties, such as a third-party system170. Location stores may be used for storing location information received fromclient systems130 associated with users. Advertisement-pricing modules may combine social information, the current time, location information, or other suitable information to provide relevant advertisements, in the form of notifications, to a user.
Assistant Systems
• FIG. 2 illustrates an example architecture of theassistant system140. In particular embodiments, theassistant system140 may assist a user to obtain information or services. Theassistant system140 may enable the user to interact with it with multi-modal user input (such as voice, text, image, video, motion) in stateful and multi-turn conversations to get assistance. Theassistant system140 may create and store a user profile comprising both personal and contextual information associated with the user. In particular embodiments, theassistant system140 may analyze the user input using natural-language understanding. The analysis may be based on the user profile for more personalized and context-aware understanding. Theassistant system140 may resolve entities associated with the user input based on the analysis. In particular embodiments, theassistant system140 may interact with different agents to obtain information or services that are associated with the resolved entities. Theassistant system140 may generate a response for the user regarding the information or services by using natural-language generation. Through the interaction with the user, theassistant system140 may use dialog management techniques to manage and forward the conversation flow with the user. In particular embodiments, theassistant system140 may further assist the user to effectively and efficiently digest the obtained information by summarizing the information. Theassistant system140 may also assist the user to be more engaging with an online social network by providing tools that help the user interact with the online social network (e.g., creating posts, comments, messages). Theassistant system140 may additionally assist the user to manage different tasks such as keeping track of events. In particular embodiments, theassistant system140 may proactively execute pre-authorized tasks that are relevant to user interests and preferences based on the user profile, at a time relevant for the user, without a user input. In particular embodiments, theassistant system140 may check privacy settings to ensure that accessing a user's profile or other user information and executing different tasks are permitted subject to the user's privacy settings. More information on assisting users subject to privacy settings may be found in U.S. Patent Application No. 62/675,090, filed 22 May 2018, which is incorporated by reference.
• In particular embodiments, theassistant system140 may receive a user input from theassistant application136 in theclient system130 associated with the user. In particular embodiments, the user input may be a user generated input that is sent to theassistant system140 in a single turn. If the user input is based on a text modality, theassistant system140 may receive it at amessaging platform205. If the user input is based on an audio modality (e.g., the user may speak to theassistant application136 or send a video including speech to the assistant application136), theassistant system140 may process it using an automatic speech recognition (ASR)module210 to convert the user input into text. If the user input is based on an image or video modality, theassistant system140 may process it using optical character recognition techniques within themessaging platform205 to convert the user input into text. The output of themessaging platform205 or theASR module210 may be received at anassistant xbot215. More information on handling user input based on different modalities may be found in U.S. patent application Ser. No. 16/053,600, filed 2 Aug. 2018, which is incorporated by reference.
• In particular embodiments, theassistant xbot215 may be a type of chat bot. Theassistant xbot215 may comprise a programmable service channel, which may be a software code, logic, or routine that functions as a personal assistant to the user. Theassistant xbot215 may work as the user's portal to theassistant system140. Theassistant xbot215 may therefore be considered as a type of conversational agent. In particular embodiments, theassistant xbot215 may send the textual user input to a natural-language understanding (NLU)module220 to interpret the user input. In particular embodiments, theNLU module220 may get information from auser context engine225 and a semantic information aggregator (SIA)230 to accurately understand the user input. Theuser context engine225 may store the user profile of the user. The user profile of the user may comprise user-profile data including demographic information, social information, and contextual information associated with the user. The user-profile data may also include user interests and preferences on a plurality of topics, aggregated through conversations on news feed, search logs,messaging platform205, etc. The usage of a user profile may be protected behind aprivacy check module245 to ensure that a user's information can be used only for his/her benefit, and not shared with anyone else. More information on user profiles may be found in U.S. patent application Ser. No. 15/967,239, filed 30 Apr. 2018, which is incorporated by reference. Thesemantic information aggregator230 may provide ontology data associated with a plurality of predefined domains, intents, and slots to theNLU module220. In particular embodiments, a domain may denote a social context of interaction, e.g., education. An intent may be an element in a pre-defined taxonomy of semantic intentions, which may indicate a purpose of a user interacting with theassistant system140. In particular embodiments, an intent may be an output of theNLU module220 if the user input comprises a text/speech input. TheNLU module220 may classify the text/speech input into a member of the pre-defined taxonomy, e.g., for the input “Play Beethoven's 5th,” theNLU module220 may classify the input as having the intent [IN:play_music]. In particular embodiments, a domain may be conceptually a namespace for a set of intents, e.g., music. A slot may be a named sub-string with the user input, representing a basic semantic entity. For example, a slot for “pizza” may be [SL:dish]. In particular embodiments, a set of valid or expected named slots may be conditioned on the classified intent. As an example and not by way of limitation, for [IN:play_music], a slot may be [SL:song_name]. Thesemantic information aggregator230 may additionally extract information from a social graph, a knowledge graph, and a concept graph, and retrieve a user's profile from theuser context engine225. Thesemantic information aggregator230 may further process information from these different sources by determining what information to aggregate, annotating n-grams of the user input, ranking the n-grams with confidence scores based on the aggregated information, formulating the ranked n-grams into features that can be used by theNLU module220 for understanding the user input. More information on aggregating semantic information may be found in U.S. patent application Ser. No. 15/967,342, filed 30 Apr. 2018, which is incorporated by reference. Based on the output of theuser context engine225 and thesemantic information aggregator230, theNLU module220 may identify a domain, an intent, and one or more slots from the user input in a personalized and context-aware manner. As an example and not by way of limitation, a user input may comprise “show me how to get to the Starbucks”. TheNLU module220 may identify the particular Starbucks that the user wants to go based on the user's personal information and the associated contextual information. In particular embodiments, theNLU module220 may comprise a lexicon of language and a parser and grammar rules to partition sentences into an internal representation. TheNLU module220 may also comprise one or more programs that perform naive semantics or stochastic semantic analysis to the use of pragmatics to understand a user input. In particular embodiments, the parser may be based on a deep learning architecture comprising multiple long-short term memory (LSTM) networks. As an example and not by way of limitation, the parser may be based on a recurrent neural network grammar (RNNG) model, which is a type of recurrent and recursive LSTM algorithm. More information on natural-language understanding may be found in U.S. patent application Ser. No. 16/011,062, filed 18 Jun. 2018, U.S. patent application Ser. No. 16/025,317, filed 2 Jul. 2018, and U.S. patent application Ser. No. 16/038,120, filed 17 Jul. 2018, each of which is incorporated by reference.
• In particular embodiments, the identified domain, intent, and one or more slots from theNLU module220 may be sent to adialog engine235. In particular embodiments, thedialog engine235 may manage the dialog state and flow of the conversation between the user and theassistant xbot215. Thedialog engine235 may additionally store previous conversations between the user and theassistant xbot215. In particular embodiments, thedialog engine235 may communicate with anentity resolution module240 to resolve entities associated with the one or more slots, which supports thedialog engine235 to forward the flow of the conversation between the user and theassistant xbot215. In particular embodiments, theentity resolution module240 may access the social graph, the knowledge graph, and the concept graph when resolving the entities. Entities may include, for example, unique users or concepts, each of which may have a unique identifier (ID). As an example and not by way of limitation, the knowledge graph may comprise a plurality of entities. Each entity may comprise a single record associated with one or more attribute values. The particular record may be associated with a unique entity identifier. Each record may have diverse values for an attribute of the entity. Each attribute value may be associated with a confidence probability. A confidence probability for an attribute value represents a probability that the value is accurate for the given attribute. Each attribute value may be also associated with a semantic weight. A semantic weight for an attribute value may represent how the value semantically appropriate for the given attribute considering all the available information. For example, the knowledge graph may comprise an entity of a movie “The Martian” (2015), which includes information that has been extracted from multiple content sources (e.g., movie review sources, media databases, and entertainment content sources), and then deduped, resolved, and fused to generate the single unique record for the knowledge graph. The entity may be associated with a space attribute value which indicates the genre of the movie “The Martian” (2015). More information on the knowledge graph may be found in U.S. patent application Ser. No. 16/048,049, filed 27 Jul. 2018, and U.S. patent application Ser. No. 16/048,101, filed 27 Jul. 2018, each of which is incorporated by reference. Theentity resolution module240 may additionally request a user profile of the user associated with the user input from theuser context engine225. In particular embodiments, theentity resolution module240 may communicate with aprivacy check module245 to guarantee that the resolving of the entities does not violate privacy policies. In particular embodiments, theprivacy check module245 may use an authorization/privacy server to enforce privacy policies. As an example and not by way of limitation, an entity to be resolved may be another user who specifies in his/her privacy settings that his/her identity should not be searchable on the online social network, and thus theentity resolution module240 may not return that user's identifier in response to a request. Based on the information obtained from the social graph, knowledge graph, concept graph, and user profile, and subject to applicable privacy policies, theentity resolution module240 may therefore accurately resolve the entities associated with the user input in a personalized and context-aware manner. In particular embodiments, each of the resolved entities may be associated with one or more identifiers hosted by the social-networking system160. As an example and not by way of limitation, an identifier may comprise a unique user identifier (ID). In particular embodiments, each of the resolved entities may be also associated with a confidence score. More information on resolving entities may be found in U.S. patent application Ser. No. 16/048,049, filed 27 Jul. 2018, and U.S. patent application Ser. No. 16/048,072, filed 27 Jul. 2018, each of which is incorporated by reference.
• In particular embodiments, thedialog engine235 may communicate with different agents based on the identified intent and domain, and the resolved entities. In particular embodiments, an agent may be an implementation that serves as a broker across a plurality of content providers for one domain. A content provider may be an entity responsible for carrying out an action associated with an intent or completing a task associated with the intent. As an example and not by way of limitation, multiple device-specific implementations (e.g., real-time calls for aclient system130 or a messaging application on the client system130) may be handled internally by a single agent. Alternatively, these device-specific implementations may be handled by multiple agents associated with multiple domains. In particular embodiments, the agents may comprise first-party agents250 and third-party agents255. In particular embodiments, first-party agents250 may comprise internal agents that are accessible and controllable by the assistant system140 (e.g. agents associated with services provided by the online social network). In particular embodiments, third-party agents255 may comprise external agents that theassistant system140 has no control over (e.g., music streams agents, ticket sales agents). The first-party agents250 may be associated with first-party providers260 that provide content objects and/or services hosted by the social-networking system160. The third-party agents255 may be associated with third-party providers265 that provide content objects and/or services hosted by the third-party system170.
• In particular embodiments, the communication from thedialog engine235 to the first-party agents250 may comprise requesting particular content objects and/or services provided by the first-party providers260. As a result, the first-party agents250 may retrieve the requested content objects from the first-party providers260 and/or execute tasks that command the first-party providers260 to perform the requested services. In particular embodiments, the communication from thedialog engine235 to the third-party agents255 may comprise requesting particular content objects and/or services provided by the third-party providers265. As a result, the third-party agents255 may retrieve the requested content objects from the third-party providers265 and/or execute tasks that command the third-party providers265 to perform the requested services. The third-party agents255 may access theprivacy check module245 to guarantee no privacy violations before interacting with the third-party providers265. As an example and not by way of limitation, the user associated with the user input may specify in his/her privacy settings that his/her profile information is invisible to any third-party content providers. Therefore, when retrieving content objects associated with the user input from the third-party providers265, the third-party agents255 may complete the retrieval without revealing to the third-party providers265 which user is requesting the content objects.
• In particular embodiments, each of the first-party agents250 or third-party agents255 may be designated for a particular domain. As an example and not by way of limitation, the domain may comprise weather, transportation, music, etc. In particular embodiments, theassistant system140 may use a plurality of agents collaboratively to respond to a user input. As an example and not by way of limitation, the user input may comprise “direct me to my next meeting.” Theassistant system140 may use a calendar agent to retrieve the location of the next meeting. Theassistant system140 may then use a navigation agent to direct the user to the next meeting.
• In particular embodiments, each of the first-party agents250 or third-party agents255 may retrieve a user profile from theuser context engine225 to execute tasks in a personalized and context-aware manner. As an example and not by way of limitation, a user input may comprise “book me a ride to the airport.” A transportation agent may execute the task of booking the ride. The transportation agent may retrieve the user profile of the user from theuser context engine225 before booking the ride. For example, the user profile may indicate that the user prefers taxis, so the transportation agent may book a taxi for the user. As another example, the contextual information associated with the user profile may indicate that the user is in a hurry so the transportation agent may book a ride from a ride-sharing service for the user since it may be faster to get a car from a ride-sharing service than a taxi company. In particular embodiment, each of the first-party agents250 or third-party agents255 may take into account other factors when executing tasks. As an example and not by way of limitation, other factors may comprise price, rating, efficiency, partnerships with the online social network, etc.
• In particular embodiments, thedialog engine235 may communicate with a conversational understanding composer (CU composer)270. Thedialog engine235 may send the requested content objects and/or the statuses of the requested services to theCU composer270. In particular embodiments, thedialog engine235 may send the requested content objects and/or the statuses of the requested services as a <k, c, u, d> tuple, in which k indicates a knowledge source, c indicates a communicative goal, u indicates a user model, and d indicates a discourse model. In particular embodiments, theCU composer270 may comprise a natural-language generator (NLG)271 and a user interface (UI)payload generator272. The natural-language generator271 may generate a communication content based on the output of thedialog engine235. In particular embodiments, theNLG271 may comprise a content determination component, a sentence planner, and a surface realization component. The content determination component may determine the communication content based on the knowledge source, communicative goal, and the user's expectations. As an example and not by way of limitation, the determining may be based on a description logic. The description logic may comprise, for example, three fundamental notions which are individuals (representing objects in the domain), concepts (describing sets of individuals), and roles (representing binary relations between individuals or concepts). The description logic may be characterized by a set of constructors that allow the natural-language generator271 to build complex concepts/roles from atomic ones. In particular embodiments, the content determination component may perform the following tasks to determine the communication content. The first task may comprise a translation task, in which the input to the natural-language generator271 may be translated to concepts. The second task may comprise a selection task, in which relevant concepts may be selected among those resulted from the translation task based on the user model. The third task may comprise a verification task, in which the coherence of the selected concepts may be verified. The fourth task may comprise an instantiation task, in which the verified concepts may be instantiated as an executable file that can be processed by the natural-language generator271. The sentence planner may determine the organization of the communication content to make it human understandable. The surface realization component may determine specific words to use, the sequence of the sentences, and the style of the communication content. TheUI payload generator272 may determine a preferred modality of the communication content to be presented to the user. In particular embodiments, theCU composer270 may communicate with theprivacy check module245 to make sure the generation of the communication content follows the privacy policies. In particular embodiments, theCU composer270 may retrieve a user profile from theuser context engine225 when generating the communication content and determining the modality of the communication content. As a result, the communication content may be more natural, personalized, and context-aware for the user. As an example and not by way of limitation, the user profile may indicate that the user likes short sentences in conversations so the generated communication content may be based on short sentences. As another example and not by way of limitation, the contextual information associated with the user profile may indicated that the user is using a device that only outputs audio signals so theUI payload generator272 may determine the modality of the communication content as audio. More information on natural-language generation may be found in U.S. patent application Ser. No. 15/967,279, filed 30 Apr. 2018, and U.S. patent application Ser. No. 15/966,455, filed 30 Apr. 2018, each of which is incorporated by reference.
• In particular embodiments, theCU composer270 may send the generated communication content to theassistant xbot215. In particular embodiments, theassistant xbot215 may send the communication content to themessaging platform205. Themessaging platform205 may further send the communication content to theclient system130 via theassistant application136. In alternative embodiments, theassistant xbot215 may send the communication content to a text-to-speech (TTS)module275. TheTTS module275 may convert the communication content to an audio clip. TheTTS module275 may further send the audio clip to theclient system130 via theassistant application136.
• In particular embodiments, theassistant xbot215 may interact with aproactive inference layer280 without receiving a user input. Theproactive inference layer280 may infer user interests and preferences based on the user profile that is retrieved from theuser context engine225. In particular embodiments, theproactive inference layer280 may further communicate withproactive agents285 regarding the inference. Theproactive agents285 may execute proactive tasks based on the inference. As an example and not by way of limitation, the proactive tasks may comprise sending content objects or providing services to the user. In particular embodiments, each proactive task may be associated with an agenda item. The agenda item may comprise a recurring item such as a daily digest. The agenda item may also comprise a one-time item. In particular embodiments, aproactive agent285 may retrieve the user profile from theuser context engine225 when executing the proactive task. Therefore, theproactive agent285 may execute the proactive task in a personalized and context-aware manner. As an example and not by way of limitation, the proactive inference layer may infer that the user likes the band Maroon5 and theproactive agent285 may generate a recommendation of Maroon5's new song/album to the user.
• In particular embodiments, theproactive agent285 may generate candidate entities associated with the proactive task based on a user profile. The generation may be based on a straightforward backend query using deterministic filters to retrieve the candidate entities from a structured data store. The generation may be alternatively based on a machine-learning model that is trained based on the user profile, entity attributes, and relevance between users and entities. As an example and not by way of limitation, the machine-learning model may be based on support vector machines (SVM). As another example and not by way of limitation, the machine-learning model may be based on a regression model. As another example and not by way of limitation, the machine-learning model may be based on a deep convolutional neural network (DCNN). In particular embodiments, theproactive agent285 may also rank the generated candidate entities based on the user profile and the content associated with the candidate entities. The ranking may be based on the similarities between a user's interests and the candidate entities. As an example and not by way of limitation, theassistant system140 may generate a feature vector representing a user's interest and feature vectors representing the candidate entities. Theassistant system140 may then calculate similarity scores (e.g., based on cosine similarity) between the feature vector representing the user's interest and the feature vectors representing the candidate entities. The ranking may be alternatively based on a ranking model that is trained based on user feedback data.
• In particular embodiments, the proactive task may comprise recommending the candidate entities to a user. Theproactive agent285 may schedule the recommendation, thereby associating a recommendation time with the recommended candidate entities. The recommended candidate entities may be also associated with a priority and an expiration time. In particular embodiments, the recommended candidate entities may be sent to a proactive scheduler. The proactive scheduler may determine an actual time to send the recommended candidate entities to the user based on the priority associated with the task and other relevant factors (e.g., clicks and impressions of the recommended candidate entities). In particular embodiments, the proactive scheduler may then send the recommended candidate entities with the determined actual time to an asynchronous tier. The asynchronous tier may temporarily store the recommended candidate entities as a job. In particular embodiments, the asynchronous tier may send the job to thedialog engine235 at the determined actual time for execution. In alternative embodiments, the asynchronous tier may execute the job by sending it to other surfaces (e.g., other notification services associated with the social-networking system160). In particular embodiments, thedialog engine235 may identify the dialog intent, state, and history associated with the user. Based on the dialog intent, thedialog engine235 may select some candidate entities among the recommended candidate entities to send to theclient system130. In particular embodiments, the dialog state and history may indicate if the user is engaged in an ongoing conversation with theassistant xbot215. If the user is engaged in an ongoing conversation and the priority of the task of recommendation is low, thedialog engine235 may communicate with the proactive scheduler to reschedule a time to send the selected candidate entities to theclient system130. If the user is engaged in an ongoing conversation and the priority of the task of recommendation is high, thedialog engine235 may initiate a new dialog session with the user in which the selected candidate entities may be presented. As a result, the interruption of the ongoing conversation may be prevented. When it is determined that sending the selected candidate entities is not interruptive to the user, thedialog engine235 may send the selected candidate entities to theCU composer270 to generate a personalized and context-aware communication content comprising the selected candidate entities, subject to the user's privacy settings. In particular embodiments, theCU composer270 may send the communication content to theassistant xbot215 which may then send it to theclient system130 via themessaging platform205 or theTTS module275. More information on proactively assisting users may be found in U.S. patent application Ser. No. 15/967,193, filed 30 Apr. 2018, and U.S. patent application Ser. No. 16/036,827, filed 16 Jul. 2018, each of which is incorporated by reference.
• In particular embodiments, theassistant xbot215 may communicate with aproactive agent285 in response to a user input. As an example and not by way of limitation, the user may ask theassistant xbot215 to set up a reminder. Theassistant xbot215 may request aproactive agent285 to set up such reminder and theproactive agent285 may proactively execute the task of reminding the user at a later time.
• In particular embodiments, theassistant system140 may comprise asummarizer290. Thesummarizer290 may provide customized news feed summaries to a user. In particular embodiments, thesummarizer290 may comprise a plurality of meta agents. The plurality of meta agents may use the first-party agents250, third-party agents255, orproactive agents285 to generated news feed summaries. In particular embodiments, thesummarizer290 may retrieve user interests and preferences from theproactive inference layer280. Thesummarizer290 may then retrieve entities associated with the user interests and preferences from theentity resolution module240. Thesummarizer290 may further retrieve a user profile from theuser context engine225. Based on the information from theproactive inference layer280, theentity resolution module240, and theuser context engine225, thesummarizer290 may generate personalized and context-aware summaries for the user. In particular embodiments, thesummarizer290 may send the summaries to theCU composer270. TheCU composer270 may process the summaries and send the processing results to theassistant xbot215. Theassistant xbot215 may then send the processed summaries to theclient system130 via themessaging platform205 or theTTS module275. More information on summarization may be found in U.S. patent application Ser. No. 15/967,290, filed 30 Apr. 2018, which is incorporated by reference.
• FIG. 3 illustrates an example diagram flow of responding to a user request by theassistant system140. In particular embodiments, theassistant xbot215 may access arequest manager305 upon receiving the user request. Therequest manager305 may comprise acontext extractor306 and a conversational understanding object generator (CU object generator)307. Thecontext extractor306 may extract contextual information associated with the user request. Thecontext extractor306 may also update contextual information based on theassistant application136 executing on theclient system130. As an example and not by way of limitation, the update of contextual information may comprise content items are displayed on theclient system130. As another example and not by way of limitation, the update of contextual information may comprise whether an alarm is set on theclient system130. As another example and not by way of limitation, the update of contextual information may comprise whether a song is playing on theclient system130. TheCU object generator307 may generate particular content objects relevant to the user request. The content objects may comprise dialog-session data and features associated with the user request, which may be shared with all the modules of theassistant system140. In particular embodiments, therequest manager305 may store the contextual information and the generated content objects indata store310 which is a particular data store implemented in theassistant system140.
• In particular embodiments, therequest manger305 may send the generated content objects to theNLU module220. TheNLU module220 may perform a plurality of steps to process the content objects. Atstep221, theNLU module220 may generate a whitelist for the content objects. In particular embodiments, the whitelist may comprise interpretation data matching the user request. Atstep222, theNLU module220 may perform a featurization based on the whitelist. Atstep223, theNLU module220 may perform domain classification/selection on user request based on the features resulted from the featurization to classify the user request into predefined domains. The domain classification/selection results may be further processed based on two related procedures. Atstep224a, theNLU module220 may process the domain classification/selection result using an intent classifier. The intent classifier may determine the user's intent associated with the user request. In particular embodiments, there may be one intent classifier for each domain to determine the most possible intents in a given domain. As an example and not by way of limitation, the intent classifier may be based on a machine-learning model that may take the domain classification/selection result as input and calculate a probability of the input being associated with a particular predefined intent. Atstep224b, the NLU module may process the domain classification/selection result using a meta-intent classifier. The meta-intent classifier may determine categories that describe the user's intent. In particular embodiments, intents that are common to multiple domains may be processed by the meta-intent classifier. As an example and not by way of limitation, the meta-intent classifier may be based on a machine-learning model that may take the domain classification/selection result as input and calculate a probability of the input being associated with a particular predefined meta-intent. Atstep225a, theNLU module220 may use a slot tagger to annotate one or more slots associated with the user request. In particular embodiments, the slot tagger may annotate the one or more slots for the n-grams of the user request. Atstep225b, theNLU module220 may use a meta slot tagger to annotate one or more slots for the classification result from the meta-intent classifier. In particular embodiments, the meta slot tagger may tag generic slots such as references to items (e.g., the first), the type of slot, the value of the slot, etc. As an example and not by way of limitation, a user request may comprise “change 500 dollars in my account to Japanese yen.” The intent classifier may take the user request as input and formulate it into a vector. The intent classifier may then calculate probabilities of the user request being associated with different predefined intents based on a vector comparison between the vector representing the user request and the vectors representing different predefined intents. In a similar manner, the slot tagger may take the user request as input and formulate each word into a vector. The intent classifier may then calculate probabilities of each word being associated with different predefined slots based on a vector comparison between the vector representing the word and the vectors representing different predefined slots. The intent of the user may be classified as “changing money”. The slots of the user request may comprise “500”, “dollars”, “account”, and “Japanese yen”. The meta-intent of the user may be classified as “financial service”. The meta slot may comprise “finance”.
• In particular embodiments, theNLU module220 may improve the domain classification/selection of the content objects by extracting semantic information from thesemantic information aggregator230. In particular embodiments, thesemantic information aggregator230 may aggregate semantic information in the following way. Thesemantic information aggregator230 may first retrieve information from theuser context engine225. In particular embodiments, theuser context engine225 may compriseoffline aggregators226 and anonline inference service227. Theoffline aggregators226 may process a plurality of data associated with the user that are collected from a prior time window. As an example and not by way of limitation, the data may include news feed posts/comments, interactions with news feed posts/comments, Instagram posts/comments, search history, etc. that are collected from a prior 90-day window. The processing result may be stored in theuser context engine225 as part of the user profile. Theonline inference service227 may analyze the conversational data associated with the user that are received by theassistant system140 at a current time. The analysis result may be stored in theuser context engine225 also as part of the user profile. In particular embodiments, both theoffline aggregators226 andonline inference service227 may extract personalization features from the plurality of data. The extracted personalization features may be used by other modules of theassistant system140 to better understand user input. In particular embodiments, thesemantic information aggregator230 may then process the retrieved information, i.e., a user profile, from theuser context engine225 in the following steps. Atstep231, thesemantic information aggregator230 may process the retrieved information from theuser context engine225 based on natural-language processing (NLP). In particular embodiments, thesemantic information aggregator230 may tokenize text by text normalization, extract syntax features from text, and extract semantic features from text based on NLP. Thesemantic information aggregator230 may additionally extract features from contextual information, which is accessed from dialog history between a user and theassistant system140. Thesemantic information aggregator230 may further conduct global word embedding, domain-specific embedding, and/or dynamic embedding based on the contextual information. Atstep232, the processing result may be annotated with entities by an entity tagger. Based on the annotations, thesemantic information aggregator230 may generate dictionaries for the retrieved information atstep233. In particular embodiments, the dictionaries may comprise global dictionary features which can be updated dynamically offline. Atstep234, thesemantic information aggregator230 may rank the entities tagged by the entity tagger. In particular embodiments, thesemantic information aggregator230 may communicate withdifferent graphs330 including social graph, knowledge graph, and concept graph to extract ontology data that is relevant to the retrieved information from theuser context engine225. In particular embodiments, thesemantic information aggregator230 may aggregate the user profile, the ranked entities, and the information from thegraphs330. Thesemantic information aggregator230 may then send the aggregated information to theNLU module220 to facilitate the domain classification/selection.
• In particular embodiments, the output of theNLU module220 may be sent to aco-reference module315 to interpret references of the content objects associated with the user request. In particular embodiments, theco-reference module315 may be used to identify an item to which the user request refers. Theco-reference module315 may comprisereference creation316 andreference resolution317. In particular embodiments, thereference creation316 may create references for entities determined by theNLU module220. Thereference resolution317 may resolve these references accurately. As an example and not by way of limitation, a user request may comprise “find me the nearest supermarket and direct me there”. Theco-reference module315 may interpret “there” as “the nearest supermarket”. In particular embodiments, theco-reference module315 may access theuser context engine225 and thedialog engine235 when necessary to interpret references with improved accuracy.
• In particular embodiments, the identified domains, intents, meta-intents, slots, and meta slots, along with the resolved references may be sent to theentity resolution module240 to resolve relevant entities. Theentity resolution module240 may execute generic and domain-specific entity resolution. In particular embodiments, theentity resolution module240 may comprisedomain entity resolution241 andgeneric entity resolution242. Thedomain entity resolution241 may resolve the entities by categorizing the slots and meta slots into different domains. In particular embodiments, entities may be resolved based on the ontology data extracted from thegraphs330. The ontology data may comprise the structural relationship between different slots/meta-slots and domains. The ontology may also comprise information of how the slots/meta-slots may be grouped, related within a hierarchy where the higher level comprises the domain, and subdivided according to similarities and differences. Thegeneric entity resolution242 may resolve the entities by categorizing the slots and meta slots into different generic topics. In particular embodiments, the resolving may be also based on the ontology data extracted from thegraphs330. The ontology data may comprise the structural relationship between different slots/meta-slots and generic topics. The ontology may also comprise information of how the slots/meta-slots may be grouped, related within a hierarchy where the higher level comprises the topic, and subdivided according to similarities and differences. As an example and not by way of limitation, in response to the input of an inquiry of the advantages of a Tesla car, thegeneric entity resolution242 may resolve a Tesla car as vehicle and thedomain entity resolution241 may resolve the Tesla car as electric car.
• In particular embodiments, the output of theentity resolution module240 may be sent to thedialog engine235 to forward the flow of the conversation with the user. Thedialog engine235 may comprisedialog intent resolution236 and dialog state update/ranker237. In particular embodiments, thedialog intent resolution236 may resolve the user intent associated with the current dialog session based on dialog history between the user and theassistant system140. Thedialog intent resolution236 may map intents determined by theNLU module220 to different dialog intents. Thedialog intent resolution236 may further rank dialog intents based on signals from theNLU module220, theentity resolution module240, and dialog history between the user and theassistant system140. In particular embodiments, the dialog state update/ranker237 may update/rank the dialog state of the current dialog session. As an example and not by way of limitation, the dialog state update/ranker237 may update the dialog state as “completed” if the dialog session is over. As another example and not by way of limitation, the dialog state update/ranker237 may rank the dialog state based on a priority associated with it.
• In particular embodiments, thedialog engine235 may communicate with atask completion module335 about the dialog intent and associated content objects. In particular embodiments, thetask completion module335 may rank different dialog hypotheses for different dialog intents. Thetask completion module335 may comprise anaction selection component336. In particular embodiments, thedialog engine235 may additionally check againstdialog policies320 regarding the dialog state. In particular embodiments, adialog policy320 may comprise a data structure that describes an execution plan of an action by anagent340. Anagent340 may select among registered content providers to complete the action. The data structure may be constructed by thedialog engine235 based on an intent and one or more slots associated with the intent. Adialog policy320 may further comprise multiple goals related to each other through logical operators. In particular embodiments, a goal may be an outcome of a portion of the dialog policy and it may be constructed by thedialog engine235. A goal may be represented by an identifier (e.g., string) with one or more named arguments, which parameterize the goal. As an example and not by way of limitation, a goal with its associated goal argument may be represented as {confirm_artist, args:{artist: “Madonna”}}. In particular embodiments, a dialog policy may be based on a tree-structured representation, in which goals are mapped to leaves of the tree. In particular embodiments, thedialog engine235 may execute adialog policy320 to determine the next action to carry out. Thedialog policies320 may comprisegeneric policy321 and domainspecific policies322, both of which may guide how to select the next system action based on the dialog state. In particular embodiments, thetask completion module335 may communicate withdialog policies320 to obtain the guidance of the next system action. In particular embodiments, theaction selection component336 may therefore select an action based on the dialog intent, the associated content objects, and the guidance fromdialog policies320.
• In particular embodiments, the output of thetask completion module335 may be sent to theCU composer270. In alternative embodiments, the selected action may require one ormore agents340 to be involved. As a result, thetask completion module335 may inform theagents340 about the selected action. Meanwhile, thedialog engine235 may receive an instruction to update the dialog state. As an example and not by way of limitation, the update may comprise awaiting agents' response. In particular embodiments, theCU composer270 may generate a communication content for the user using theNLG271 based on the output of thetask completion module335. In particular embodiments, theNLG271 may use different language models and/or language templates to generate natural language outputs. The generation of natural language outputs may be application specific. The generation of natural language outputs may be also personalized for each user. TheCU composer270 may also determine a modality of the generated communication content using theUI payload generator272. Since the generated communication content may be considered as a response to the user request, theCU composer270 may additionally rank the generated communication content using aresponse ranker273. As an example and not by way of limitation, the ranking may indicate the priority of the response.
• In particular embodiments, the output of theCU composer270 may be sent to aresponse manager325. Theresponse manager325 may perform different tasks including storing/updating thedialog state326 retrieved fromdata store310 and generatingresponses327. In particular embodiments, the output ofCU composer270 may comprise one or more of natural-language strings, speech, or actions with parameters. As a result, theresponse manager325 may determine what tasks to perform based on the output ofCU composer270. In particular embodiments, the generated response and the communication content may be sent to theassistant xbot215. In alternative embodiments, the output of theCU composer270 may be additionally sent to theTTS module275 if the determined modality of the communication content is audio. The speech generated by theTTS module275 and the response generated by theresponse manager325 may be then sent to theassistant xbot215.
Gesture-Input Workflow
• FIG. 4 illustrates an example workflow of processing a user input comprising a gesture-input. A gesture-input may be an input which is based on a movement of part of the body, especially a hand or the head, to express an idea or meaning. The gesture-inputs may be in the form of image information, video information, motion information, or any combination thereof. In particular embodiments, theassistant system140 may receive auser input405 by a user from aclient system130. Theassistant system140 may send theuser input405 to an intent-understandingmodule410. The intent-understandingmodule410 may comprise a natural-language understanding (NLU)module220 and a gesture-classification model415. The gesture-classification model415 may be a machine-learning model trained offline to recognize different categories of gestures performed by users. As an example and not by way of limitation, agesture classification model415 may be based on one or more of convolutional neural networks, tensor flow, or hidden Markov models. Based on theuser input405, the intent-understandingmodule410 may use different components to determine an intent420 associated with theuser input405. If theuser input405 comprises a text-input or a speech-input, the intent-understandingmodule410 may use theNLU module220 to determine the intent420. If theuser input405 comprises a gesture-input, the intent-understandingmodule410 may use the gesture-classification model415 to determine the intent420. Thedetermined intent420 may be associated with a confidence score indicating how confident theassistant system140 is in determining the user'sintent420. In particular embodiments, the confidence score may be determined based on how closely the user's text-input or speech-input matches a known input for a givenintent420. As an example and not by way of limitation, the closeness may be based on string similarity between the text-input and the known input (text). Theassistant system140 may then send the intent420 with its confidence score to thedialog engine235. If the confidence score is above a threshold score, thedialog engine235 may determine one ormore tasks425 corresponding to the intent420 and send them to either 1st-party agents250 or 3rd-party agents255 for executing thetasks425. The execution results430 may be sent to theCU composer270. If the confidence score is below a threshold score, thedialog engine235 may determine one or more suggestedinputs435 for the user, which may help theassistant system140 determine the user's intent with a higher confidence score. Thedialog engine235 may send the suggestedinputs435 to theCU composer270. Based on the execution results430 or the suggestedinputs435, theCU composer270 may generate aresponse440. Theresponse440 may be in different modalities comprising one or more of a text, an image, a video, or an animation of a gesture. TheCU composer270 may further send theresponse440 to theclient system130. Although this disclosure describes processing a user input via particular systems in particular manners, this disclosure contemplates describes processing any suitable user input via any suitable system in any suitable manner.
Auto-Completion for Multi-Modal User Input
• In particular embodiments, theassistant system140 may receive an initial input in a first modality executed by a user as an interaction with an assistant-basedclient system130, e.g., a gesture-input to a virtual reality (VR) headset or augment reality (AR) smart glasses, determine candidate continuation-inputs in an auto-completion manner based on the initial input, and recommend these candidate continuation-inputs to the user in one or more second modalities (e.g., text), for which the user may select one of them to trigger the execution of aparticular task425. The initial input and continuation-inputs may be based on any suitable modality including text, speech, image, video, motion, orientation, etc. In addition, the modality of the initial input and the modalities of the continuation-inputs may be different. As an example and not by way of limitation, a user visiting Paris who wears AR glasses may be looking at Eiffel Tower. Based on the gaze input, theassistant system140 may suggest a gesture displayed on the virtual screen for taking a picture or suggest the user to speak “take a picture.” The user may perform the suggested gesture or speak the sentence to allow theassistant system140 to execute the task of taking a picture of Eiffel Tower. As a result, theassistant system140 may have the ability to process auser input405 in one modality and generate suggested inputs in one or more other modalities, which may be referred as auto-completion for multi-modal user input. Although this disclosure describes generating particular auto-completion for particular multi-modal user inputs via particular systems in particular manners, this disclosure contemplates generating any suitable auto-completion for any suitable multi-modal user input via any suitable system in any suitable manner.
• In particular embodiments, theassistant system140 may receive, from aclient system130 associated with a first user, an initial input from the first user. The initial input may be in a first modality. In particular embodiments, theassistant system140 may determine, by an intent-understanding module, one ormore intents420 corresponding to the initial input. Theassistant system140 may then generate, based on the one ormore intents420, one or more candidate continuation-inputs. The one or more candidate continuation-inputs may be in one or more candidate modalities, respectively. The candidate modalities may be different from the first modality. In particular embodiments, theassistant system140 may further send, to theclient system130, instructions for presenting one or more suggested inputs corresponding to one or more of the candidate continuation-inputs.
• In particular embodiments, the first modality may comprise one of audio, text, image, video, motion, or orientation. As an example and not by way of limitation, the first modality may comprise motion, and correspondingly the initial input may comprise a gesture. As another example and not by way of limitation, the first modality may comprise orientation, and correspondingly the initial input may comprise a gaze on an object. In particular embodiments, the first modality and the candidate modalities for the candidate continuation-inputs may be different. As an example and not by way of limitation, a user may be wearing AR glasses and his/her initial input may be a framing gesture by two hands for locating an object in front of him/her. Theassistant system140 may suggest the user to say “click” as a candidate continuation-input which is displayed as text on the screen of the AR glasses. Once the user says “click” theassistant system140 may further execute the task of taking a picture of the object in the frame. As a result, theassistant system140 may have technical advantage of enriching user experience with theassistant system140 as a user may interact with theassistant system140 with inputs in various modalities. Although this disclosure describes particular modalities in particular manners, this disclosure contemplates any suitable modality in any suitable manner.
• In particular embodiments, theassistant system140 may determine the candidate continuation-inputs based on the intent of the initial input and the potential entities relevant to theintent420. Theassistant system140 may first identify one or more entities associated with the one ormore intents420. Theassistant system140 may then generate the one or more candidate continuation-inputs further based the one or more entities. As an example and not by way of limitation, a user may perform a calling gesture as an initial input. Theassistant system140 may determine a calling intent and possible entities (e.g., contacts) associated with the calling intent. The entities may be suggested to the user as text and theassistant system140 may additionally ask the user to say the name of the entity that the user intended to call. Determining the candidate continuation-inputs based on theintent420 of the initial input and the potential entities relevant to the intent420 may be an effective solution for addressing the technical challenge of determining candidate continuation-inputs based on the initial input, as the intent420 and the potential entities may be necessary to determine what continuation-input a user may use to interact with theassistant system140 to execute tasks that based on them. Although this disclosure describes determining particular inputs in particular manners, this disclosure contemplates determining any suitable input in any suitable manner.
• In particular embodiments, theassistant system140 may determine that the first user needs one or more suggested inputs. Theassistant system140 may determine whether to suggest candidate continuation-inputs based on different factors. In particular embodiments, determining that the first user needs one or more suggested inputs may be based on a wake-up input from the first user. The wake-up input may comprise one or more of a voice utterance, a character string, an image, a video clip, a gesture, or a gaze. As an example and not by way of limitation, the wake-up input may be a voice command saying “assistant!” or a gesture such as snapping fingers. Once theassistant system140 receives a wake-up input, it may initiate the process for suggesting candidate continuation-inputs. Theassistant system140 may be constantly in an active status for suggesting candidate continuation-inputs. In addition, whether to suggest continuation-inputs may be based on user attention, i.e., gaze information, detected by theassistant system140. In particular embodiments, the initial input may comprise a gaze on an object and determining that the first user needs one or more suggested inputs may be further based on the gaze on the object. On the other hand, generating the one or more candidate continuation-inputs may be further based on the object. Determining the candidate continuation-inputs based on the object associated with the initial input may be an effective solution for addressing the technical challenge of determining candidate continuation-inputs based on the initial input, as the object may be necessary to determine what continuation-input a user may use to interact with the assistant system to execute tasks that based on it. As an example and not by way of limitation, if a user who wears AR glasses is staring at a pair of shoes at a store, theassistant system140 may determine that the user needs suggested inputs and generate suggested inputs such as “taking a picture” and “checking the reviews,” etc. The suggested inputs may be based on any suitable modality. For example, the suggested inputs may be presented as gestures displayed in the screen of the AR glasses, which the user can follow to perform. As another example, the suggested inputs may be audio-based, which the user can listen to. As another example and not by way of limitation, a user is having a video call on the AR glasses and then the user turned around and looked at a TV for a threshold amount of time. Theassistant system140 may then determine that the user needs suggested inputs. Theassistant system140 may present a suggested input that instructs the user to point to the TV if the user wants the video call to be mirrored to the TV. In particular embodiments, determining that the first user needs one or more suggested inputs may be further based on contextual information associated with the initial input. In particular embodiments, contextual information may comprise one or more of temporal information, location information, presence information, or social information. As an example and not by way of limitation, the user may input only a few letters and then stopped inputting text for over a certain amount of time. Theassistant system140 may then determine that the user needs suggested inputs as the user may be having difficulty inputting the intended text. As another example and not by way of limitation, the user may be at a self-service restaurant and send an image of a dish to theassistant system140. Theassistant system140 may then determine that user needs suggested inputs such as the name of the dish or a request to order that dish. In particular embodiments, determining that the first user needs one or more suggested inputs may be further based on the one ormore intents420. As an example and not by way of limitation, a user's initial input may indicate that the user wants to buy a movie ticket but the user did not provide any additional information. Theassistant system140 may then determine that the user needs suggested inputs such as names of recent popular movies. Although this disclosure describes determining whether to suggest particular inputs in particular manners, this disclosure contemplates determining whether to suggest any suitable input in any suitable manner.
• In particular embodiments, theassistant system140 may further receive, from theclient system130, a user-selected input from the first user. The user-selected input may comprise one of the suggested inputs. Theassistant system140 may then execute one ormore tasks425 based on the user-selected input. In particular embodiments, theassistant system140 may suggest candidate continuation-inputs for more than once. In other words, theassistant system140 may guide the user through a sequence of candidate continuation-inputs, in which each of the candidate continuation-inputs may be mapped to aspecific intent420. To be more specific, theassistant system140 may receive, from theclient system130, a first user-selected input from the first user. The first user-selected input may comprise one of the suggested inputs and the first user-selected input may be associated with afirst intent420. Theassistant system140 may then generate, based on the first user-selected input, one or more additional candidate continuation-inputs. Each of the one or more additional candidate continuation-inputs may be associated with thefirst intent420. Theassistant system140 may then send, to theclient system130, instructions for presenting one or more additional suggested inputs corresponding to one or more of the additional candidate continuation-inputs. Theassistant system140 may then receive, from theclient system130, a second user-selected input from the first user. The second user-selected input may comprise one of the additional suggested inputs. Theassistant system140 may further execute one ormore tasks425 based on the second user-selected input. The one ormore tasks425 may correspond to thefirst intent420. In particular embodiments, theassistant system140 may continue the aforementioned process until one ormore tasks425 that require a whole sequence of inputs are executed. As an example and not by way of limitation, a user wearing AR glasses may want to set a reminder by performing a “reminder” gesture. Theassistant system140 may recognize such gesture and generate suggestions of what events the reminder is for. Theassistant system140 may present three suggested events in text, e.g., “mow the lawn”, “feed the dog”, and “buy a gift for Mom”. The user may tap “feeding the dog”. Theassistant system140 may further generate suggested time (e.g., 10 am, 3 pm, and 7 pm) and ask the user to select the time either by tapping or speaking. After the user selects the time (e.g., by speaking “7 pm”), theassistant system140 may set the reminder accordingly. As a result, theassistant system140 may have a technical advantage of increasing the degree of users engaging with theassistant system140 by guiding users with sequences of suggested inputs to explore various tasks that the users may request assistance. Although this disclosure describes generating particular sequences of inputs in particular manners, this disclosure contemplates generating any suitable sequence of input in any suitable manner.
Auto-Completion for Gesture-Input
• In particular embodiments, theassistant system140 may receive an incomplete gesture executed by a user as an interaction with an assistant-basedclient system130, e.g., a VR headset or AR smart glasses, determine candidate gestures in an auto-completion manner based on the incomplete gesture, recommend these candidate gestures to the user, for which the user may select one of them to trigger the execution of aparticular task425. As an example and not by way of limitation, a user wearing AR glasses may not know what gesture to perform to trigger a particular function of theassistant system140. The user may start moving his/her hand but pause in the air. Accordingly, theassistant system140 may analyze the incomplete gesture of the user and determine possible candidate gestures. Theassistant system140 may display these candidate gestures visually via the AR glasses to the user. After the user selects one of them, theassistant system140 may further execute atask425 corresponding to that gesture. In summary, theassistant system140 may use gesture recognition techniques on incomplete gestures to determine the user-intended gestures and suggest the determined gestures as guidance to the user, which may be referred as auto-completion for gesture-input. Although this disclosure describes generating particular auto-completion for particular gesture-inputs via particular systems in particular manners, this disclosure contemplates generating any suitable auto-completion for any suitable gesture-input via any suitable system in any suitable manner.
• In particular embodiments, theassistant system140 may receive, from a client system associated with a first user, auser input405 from the first user. Theuser input405 may comprise an incomplete gesture performed by the first user. In particular embodiments, theassistant system140 may calculate, by an intent-understanding module, one or more confidences scores for one ormore intents420 corresponding to the incomplete gesture. Theassistant system140 may then determine that the calculated confidence scores associated with each of theintents420 are below a threshold score. In particular embodiments, theassistant system140 may select, based on a personalized gesture-recognition model, one or more candidate gestures from a plurality of pre-defined gestures responsive to determining that the calculated confidence scores for each of theintents420 are below the threshold score. Each of the candidate gestures may be associated with a confidence score representing a likelihood the first user intended to input the respective candidate gesture. In particular embodiments, theassistant system140 may further send, to theclient system130, instructions for presenting one or more suggested inputs corresponding to one or more of the candidate gestures.
• In particular embodiments, theassistant system140 may recognize an incomplete gesture of auser input405. An incomplete gesture may comprise the onset of a gesture, which may indicate that the user has not completed the whole gesture yet. As an example and not by way of limitation, an open hand pausing in the air may be an incomplete gesture for the waving gesture. In particular embodiments, theassistant system140 may suggest a list of pre-defined gestures, one of which may be the gesture that the user intended to perform. As an example and not by way of limitation, each pre-defined gesture may comprise one or more of pointing, poking, tapping, waving, or swiping. As a result, theassistant system140 may have a technical advantage of improving user experience with theassistant system140 as theassistant system140 may understand a user'sintent420 based on an incomplete gesture and teach a user to complete the incomplete gesture to interact with theassistant system140. Although this disclosure describes particular gestures in particular manners, this disclosure contemplates any suitable gesture in any suitable manner.
• In particular embodiments, theassistant system140 may determine a user's intended gesture based on similarity evaluation with respect to a list of pre-defined gestures. Theassistant system140 may calculate, for each of the one or more candidate gestures, a similarity level of the candidate gesture with respect to the incomplete gesture. As an example and not by way of limitation, if a user's hand is in a particular position, theassistant system140 may determine how similar this incomplete gesture is to the pre-defined gestures. After determining which pre-defined gestures this uncompleted gesture is most similar to, theassistant system140 may select them as candidate gestures. In particular embodiments, theassistant system140 may determine whether to present candidate gestures to a user based on similarity levels. As an example and not by way of limitation, if the similarity levels for all the candidate gestures with respect to the incomplete gesture are below a pre-defined threshold level, theassistant system140 may present them to the user for selection. As another example and not by way of limitation, if two or more candidate gestures have similar similarity levels which are greater than the pre-defined threshold level, theassistant system140 may still present them to the user for disambiguation. In particular embodiments, selecting the one or more candidate gestures may be further based on the one ormore intents420. As an example and not by way of limitation, theintents420 may comprise locating an object or interacting with another user on social media. Correspondingly, the candidate gestures may comprise pointing or poking. In particular embodiments, theassistant system140 may further suggest inputs which may guide the user to complete the candidate gestures. Although this disclosure describes determining particular candidate gestures in particular manners, this disclosure contemplates determining any suitable candidate gesture in any suitable manner.
• In particular embodiments, theassistant system140 may determine the similarity level in different ways. In particular embodiments, the similarly level of each candidate gesture with respect to the incomplete gesture may be based on a trajectory of the incomplete gesture with respect to the client system. As an example and not by way of limitation, if a user moved his/her hand from left to right and paused, most likely theassistant system140 may determine some gestures that continue the motion in a relatively similar trajectory, e.g., swiping. But if the user moved his/her hand from the bottom to the top, the trajectory is different, for which theassistant system140 may determine different gestures, e.g., pointing. In particular embodiments, the similarly level of each candidate gesture with respect to the incomplete gesture may be based on an orientation of the incomplete gesture with respect to theclient system130. As an example and not by way of limitation, if the orientation is parallel to theclient system130, the gestures of waving or swiping may have higher similarity level with respect to the incomplete gesture. As another example and not by way of limitation, if the orientation is perpendicular to theclient system130, the gestures of pointing or poking may have higher similarity level with respect to the incomplete gesture. In particular embodiments, the similarly level of each candidate gesture with respect to the incomplete gesture may be based on an object associated with the incomplete gesture. As an example and not by way of limitation, the determined gestures corresponding to a phone may be different from those corresponding to a TV. In particular embodiments, the similarly level of each candidate gesture with respect to the incomplete gesture may be based on contextual information associated with the incomplete gesture. As an example and not by way of limitation, if the contextual information indicates that the user is browsing another user's social media profile, the candidate gesture may be more likely poking rather than pointing. In particular embodiments, the similarly level of each candidate gesture with respect to the incomplete gesture may be based on a position of the incomplete gesture with respect to theclient system130. As an example and not by way of limitation, if the incomplete gesture is very close to theclient system130, the candidate gesture may be more likely tapping. One technical challenge includes determining similarity levels of candidate gestures with respect to the incomplete gesture. Determining the similarity levels based on different factors including trajectory, orientation, objects, contextual information, and position associated with the incomplete gesture may be an effective solution for addressing the technical challenge of determining similarity levels of candidate gestures with respect to the incomplete gesture, as such information may provide different informative cues for theassistant system140 to calculate similarity levels. Although this disclosure describes determining particular similarity levels in particular manners, this disclosure contemplates determining any suitable similarity level in any suitable manner.
• In particular embodiments, theassistant system140 may determine whether the first user needs suggestions to complete the incomplete gesture based on the calculated confidence scores for theintents420 corresponding to the incomplete gesture. If the confidence scores are below a threshold score, it may indicate that theassistant system140 cannot determine the user'sintent420. Such difficulty in determining the user'sintent420 may be caused by the fact that the incomplete gesture is uninformative, which further indicates that the user does not know how to perform the intended gesture and may need suggestions. In particular embodiments, the threshold score may be based on a wake-up gesture performed by the first user. The wake-up gesture may indicate that the user clearly needs suggestions. As a result, theassistant system140 may set a high threshold score, which results in the calculated confidence scores being below the threshold score. Theassistant system140 may then generate candidate gestures accordingly. In summary, once a user has made a wake-up gesture, theassistant system140 may initiate the process for gesture suggestion. The assistant system may be constantly in an active status for gesture suggestion. In particular embodiments, calculating the one or more confidence scores for the one ormore intents420 corresponding to the incomplete gesture may be based on temporal information associated with the incomplete gesture. The temporal information may comprise a pause in theuser input405. As an example and not by way of limitation, the calculated confidence scores may be low when there is a pause in theuser input405. Accordingly, theassistant system140 may determine that the user needs suggestions to complete the incomplete gesture. In particular embodiments, calculating the one or more confidence scores for the one ormore intents420 corresponding to the incomplete gesture may be based on a velocity associated with the incomplete gesture. As an example and not by way of limitation, theassistant system140 may monitor how fast a user's movement is. If the user is moving fairly confident and fast the user probably already knew the gesture that he/she wants to perform, for which theassistant system140 may calculate high confidence scores and not suggest any gestures. If theassistant system140 detects that the user is hesitant of performing a gesture (e.g., moving slow), theassistant system140 may infer that the user is trying to figure out what gestures he/she can do, calculate low confidence scores, and suggest gestures accordingly. In particular embodiments, determining the first user needs suggestions to complete the incomplete gesture may be further based on patterns of the incomplete gesture. As an example and not by way of limitation, if a user is repeating the same incomplete gesture, theassistant system140 may determine a suggested gesture to complete the user's intended gesture. Determining whether to suggest candidate gestures based on different factors including confidence scores of intents, wake-up gestures, temporal information of the incomplete gesture, velocity of the incomplete gesture, and patterns of the incomplete gesture may be an effective solution for addressing the technical challenge of determining that a user needs suggested inputs for gestures, as such information may reveal a user's requirement for assistance from different perspectives. Although this disclosure describes determining whether a user needs particular suggested gestures in particular manners, this disclosure contemplates determining whether a user needs any suitable suggested gestures in any suitable manner.
• In particular embodiments, theassistant system140 may receive, from theclient system130, a user-selected input from the first user. The user-selected input may comprise one of the suggested inputs. Theassistant system140 may then execute one or more tasks based on the user-selected input. In particular embodiments, theassistant system140 may suggest candidate gestures for more than once. In other words, theassistant system140 may guide the user through a sequence of candidate gestures, in which each of the candidate gestures may be mapped to aspecific intent420. To be more specific, theassistant system140 may receive, from theclient system130, a first user-selected input from the first user. The first user-selected input may comprise one of the suggested inputs and the first user-selected input may be associated with afirst intent420. Theassistant system140 may then generate, based on the first user-selected input, one or more additional candidate gestures. Each of the one or more additional candidate gestures may be associated with thefirst intent420. Theassistant system140 may then send, to theclient system130, instructions for presenting one or more additional suggested inputs corresponding to one or more of the additional candidate gestures. Theassistant system140 may then receive, from theclient system130, a second user-selected input from the first user. The second user-selected input may comprise one of the additional suggested inputs. Theassistant system140 may further execute one ormore tasks425 based on the second user-selected input. The one ormore tasks425 may correspond to thefirst intent420. In particular embodiments, theassistant system140 may continue the aforementioned process until one ormore tasks425 that require a whole sequence of gestures are executed. As an example and not by way of limitation, a user wearing AR glasses may want to make a call and have performed an incomplete gesture. Theassistant system140 may suggest a few candidate gestures including the “call” gesture. The user may then complete the “call” gesture based on the suggestion. Theassistant system140 may then display three different people (e.g., names or photos) on the virtual screen of the VR glasses and suggest additional gestures (e.g., pointing or tapping) to the user. Accordingly, the user may perform one of the additional suggested gestures on someone he/she wants to call. Finally, theassistant system140 may execute thetask425 of calling the intended person. As another example and not by way of limitation, a user may want to take a picture of a friend and have performed an incomplete gesture. Theassistant system140 may suggest a few candidate gestures including the “taking-pictures” gesture to the user. The user may then complete the “taking-pictures” gesture based on the suggestion. After the picture is taken, theassistant system140 may further suggest some additional gestures that the user could perform on the picture. For example, one gesture may be a gesture for sending this picture to the friend. The user may thus perform such gesture to finish a sequence of gestures and theassistant system140 may send the picture to the friend subsequently. As a result, theassistant system140 may have a technical advantage of increasing the degree of users engaging with theassistant system140 by guiding users with sequences of suggested gestures to explore various tasks that the users may request assistance. Although this disclosure describes suggesting particular sequence of gestures in particular manners, this disclosure contemplates suggesting any suitable sequence of gestures in any suitable manner.
• In particular embodiments, determining candidate gestures responsive to a user's incomplete gesture may be applied to user education in assistant-based AR/VR systems. As an example and not by way of limitation, when a user initially puts on the AR glasses he may not know what gestures he can use to interact with theassistant system140. The user may raise his/her hand wondering what to do. In this case, theassistant system140 may analyze the user's current incomplete gesture and determine what gestures may be similar gestures that the user can perform. Accordingly, theassistant system140 may suggest the determined gestures by displaying them virtually on the screen of the AR glasses. In particular embodiments, determining candidate gestures responsive to a user's incomplete gesture may be suitable when only gestures are available for interacting with theassistant system140. As an example and not by way of limitation, a user cannot speak and does not have a keyboard for texting. In this case, if the user starts his/her interaction with theassistant system140 with an incomplete gesture, theassistant system140 may suggest the user to perform one or more gestures to complete aparticular task425. Although this disclosure describes particular applications of the embodiments disclosed herein in particular manners, this disclosure contemplates any suitable application of the embodiments disclosed herein in any suitable manner.
• FIG. 5 illustrates an example scenario of auto-completion formulti-modal user input405. As displayed inFIG. 5, auser500 may be wearing a pair ofAR glasses505. TheAR glasses505 may function as aclient system130 for theuser500 to interact with theassistant system140. Theuser500 may provide an initial input, which may be a “call”gesture510. Theassistant system140 may determine the user'sintent420 to call a contact corresponding to the “call”gesture510. Theassistant system140 may identify two contacts that theuser500 may want to call, e.g., his Mom and his wife Jane. Based on the identified contacts, theassistant system140 may generate two candidate continuation-inputs including selecting a contact by voice and selecting a contact by gesture. Theassistant system140 may further generate suggested inputs which may be displayed on thescreen515 of theVR glasses505. InFIG. 5, the suggested inputs are presented as “who would you like to reach? You may say ‘Mom’ or ‘Jane’ or you may tap on the photo”520. In addition, a photo of “Mom”525 and a photo of “Jane”530 may be presented in association with the suggested inputs. Although this disclosure describes particular scenarios of auto-completion for particular multi-modal user input in particular manners, this disclosure contemplates any suitable scenario of auto-completion for any suitable multi-modal user input in any suitable manner.
• FIG. 6 illustrates an example scenario of auto-completion for gesture-input. As displayed inFIG. 6, auser500 may be interacting with theassistant system140 via aclient system130. Theuser500 may want to perform a gesture to interact with theassistant system140 but does not really know what gestures he could perform. Therefore, theuser500 may hold his hand in the air, which may have formed anincomplete gesture605. Based on theincomplete gesture605, theassistant system140 may determine one or more candidate gestures using a personalized gesture-recognition model. As an example and not by way of limitation, the one or more candidate gestures may comprise two gestures, i.e., waving and tapping. Theassistant system140 may further generate suggested inputs which may be displayed on thescreen610 of theclient system130. InFIG. 6, the suggested inputs are presented as “please perform one of the following gestures to let me know how I can help you”615. In addition, an animation of the wavinggesture620 and an animation of the tappinggesture625 may be presented in association with the suggested inputs. Although this disclosure describes particular scenarios of auto-completion for particular gesture-input in particular manners, this disclosure contemplates any suitable scenario of auto-completion for any suitable gesture-input in any suitable manner.
• FIGS. 7A-7B illustrate another example scenario of auto-completion for gesture-input.FIG. 7A illustrates an example scenario of an incomplete gesture in AR/VR setting. As displayed inFIG. 7A, auser500 may be wearing a pair ofAR glasses505. TheAR glasses505 may function as aclient system130 for theuser500 to interact with theassistant system140. Theuser500 may see a dog on thescreen705 of theAR glasses505. Theuser500 may want to perform a gesture to interact with theassistant system140 but does not really know what gestures he could perform. Therefore, theuser500 may hold his hands in the air, which may have formed anincomplete gesture710. Based on theincomplete gesture710 and the focus of the user's attention (i.e., the dog), theassistant system140 may determine one or more candidate gestures using a personalized gesture-recognition model. Theassistant system140 may further generate suggested inputs which may be displayed on thescreen705 of theAR glasses505.FIG. 7B illustrates an example scenario of suggested gesture-input in AR/VR setting. InFIG. 7B, the suggested inputs are presented as “please perform the displayed gesture to take a picture”715. In addition, a “taking a picture” gesture720 (i.e., in dotted lines) may be presented in association with the suggested inputs. Although this disclosure describes particular scenarios of auto-completion for particular gesture-inputs in particular manners, this disclosure contemplates any suitable scenario of auto-completion for any suitable gesture-input in any suitable manner.
• FIG. 8 illustrates anexample method800 for suggestingmulti-modal user input405 for auto-completion. The method may begin atstep810, where theassistant system140 may receive, from aclient system130 associated with a first user, an initial input from the first user, wherein the initial input is in a first modality. Atstep820, theassistant system140 may determine, by an intent-understandingmodule410, one ormore intents420 corresponding to the initial input. Atstep830, theassistant system140 may generate, based on the one ormore intents420, one or more candidate continuation-inputs, where the one or more candidate continuation-inputs are in one or more candidate modalities, respectively, and wherein the candidate modalities are different from the first modality. Atstep840, theassistant system140 may send, to theclient system130, instructions for presenting one or more suggested inputs corresponding to one or more of the candidate continuation-inputs. Particular embodiments may repeat one or more steps of the method ofFIG. 8, where appropriate. Although this disclosure describes and illustrates particular steps of the method ofFIG. 8 as occurring in a particular order, this disclosure contemplates any suitable steps of the method ofFIG. 8 occurring in any suitable order. Moreover, although this disclosure describes and illustrates an example method for suggesting multi-modal user input for auto-completion, including the particular steps of the method ofFIG. 8, this disclosure contemplates any suitable method for suggesting multi-modal user input for auto-completion, including any suitable steps, which may include all, some, or none of the steps of the method ofFIG. 8, where appropriate. Furthermore, although this disclosure describes and illustrates particular components, devices, or systems carrying out particular steps of the method ofFIG. 8, this disclosure contemplates any suitable combination of any suitable components, devices, or systems carrying out any suitable steps of the method ofFIG. 8.
• FIG. 9 illustrates anexample method900 for suggesting candidate gestures for auto-completion. The method may begin atstep910, where theassistant system140 may receive, from aclient system130 associated with a first user, auser input405 from the first user, wherein theuser input405 comprises an incomplete gesture performed by the first user. Atstep920, theassistant system140 may calculate, by an intent-understandingmodule410, one or more confidence scores for one ormore intents420 corresponding to the incomplete gesture. Atstep930, theassistant system140 may determine that the calculated confidence scores associated with each of theintents420 are below a threshold score. Atstep940, theassistant system140 may select, based on a personalized gesture-recognition model, one or more candidate gestures from a plurality of pre-defined gestures responsive to determining that the calculated confidence scores for each of theintents420 are below the threshold score, wherein each of the candidate gestures is associated with a confidence score representing a likelihood the first user intended to input the respective candidate gesture. Atstep950, theassistant system140 may send, to theclient system130, instructions for presenting one or more suggested inputs corresponding to one or more of the candidate gestures. Particular embodiments may repeat one or more steps of the method ofFIG. 9, where appropriate. Although this disclosure describes and illustrates particular steps of the method ofFIG. 9 as occurring in a particular order, this disclosure contemplates any suitable steps of the method ofFIG. 9 occurring in any suitable order. Moreover, although this disclosure describes and illustrates an example method for suggesting candidate gestures for auto-completion, including the particular steps of the method ofFIG. 9, this disclosure contemplates any suitable method for suggesting candidate gestures for auto-completion, including any suitable steps, which may include all, some, or none of the steps of the method ofFIG. 9, where appropriate. Furthermore, although this disclosure describes and illustrates particular components, devices, or systems carrying out particular steps of the method ofFIG. 9, this disclosure contemplates any suitable combination of any suitable components, devices, or systems carrying out any suitable steps of the method ofFIG. 9.
Social Graphs
• FIG. 10 illustrates an examplesocial graph1000. In particular embodiments, the social-networking system160 may store one or moresocial graphs1000 in one or more data stores. In particular embodiments, thesocial graph1000 may include multiple nodes—which may includemultiple user nodes1002 ormultiple concept nodes1004—andmultiple edges1006 connecting the nodes. Each node may be associated with a unique entity (i.e., user or concept), each of which may have a unique identifier (ID), such as a unique number or username. The examplesocial graph1000 illustrated inFIG. 10 is shown, for didactic purposes, in a two-dimensional visual map representation. In particular embodiments, a social-networking system160, aclient system130, anassistant system140, or a third-party system170 may access thesocial graph1000 and related social-graph information for suitable applications. The nodes and edges of thesocial graph1000 may be stored as data objects, for example, in a data store (such as a social-graph database). Such a data store may include one or more searchable or queryable indexes of nodes or edges of thesocial graph1000.
• In particular embodiments, auser node1002 may correspond to a user of the social-networking system160 or theassistant system140. As an example and not by way of limitation, a user may be an individual (human user), an entity (e.g., an enterprise, business, or third-party application), or a group (e.g., of individuals or entities) that interacts or communicates with or over the social-networking system160 or theassistant system140. In particular embodiments, when a user registers for an account with the social-networking system160, the social-networking system160 may create auser node1002 corresponding to the user, and store theuser node1002 in one or more data stores. Users anduser nodes1002 described herein may, where appropriate, refer to registered users anduser nodes1002 associated with registered users. In addition or as an alternative, users anduser nodes1002 described herein may, where appropriate, refer to users that have not registered with the social-networking system160. In particular embodiments, auser node1002 may be associated with information provided by a user or information gathered by various systems, including the social-networking system160. As an example and not by way of limitation, a user may provide his or her name, profile picture, contact information, birth date, sex, marital status, family status, employment, education background, preferences, interests, or other demographic information. In particular embodiments, auser node1002 may be associated with one or more data objects corresponding to information associated with a user. In particular embodiments, auser node1002 may correspond to one or more web interfaces.
• In particular embodiments, aconcept node1004 may correspond to a concept. As an example and not by way of limitation, a concept may correspond to a place (such as, for example, a movie theater, restaurant, landmark, or city); a website (such as, for example, a website associated with the social-networking system160 or a third-party website associated with a web-application server); an entity (such as, for example, a person, business, group, sports team, or celebrity); a resource (such as, for example, an audio file, video file, digital photo, text file, structured document, or application) which may be located within the social-networking system160 or on an external server, such as a web-application server; real or intellectual property (such as, for example, a sculpture, painting, movie, game, song, idea, photograph, or written work); a game; an activity; an idea or theory; another suitable concept; or two or more such concepts. Aconcept node1004 may be associated with information of a concept provided by a user or information gathered by various systems, including the social-networking system160 and theassistant system140. As an example and not by way of limitation, information of a concept may include a name or a title; one or more images (e.g., an image of the cover page of a book); a location (e.g., an address or a geographical location); a website (which may be associated with a URL); contact information (e.g., a phone number or an email address); other suitable concept information; or any suitable combination of such information. In particular embodiments, aconcept node1004 may be associated with one or more data objects corresponding to information associated withconcept node1004. In particular embodiments, aconcept node1004 may correspond to one or more web interfaces.
• In particular embodiments, a node in thesocial graph1000 may represent or be represented by a web interface (which may be referred to as a “profile interface”). Profile interfaces may be hosted by or accessible to the social-networking system160 or theassistant system140. Profile interfaces may also be hosted on third-party websites associated with a third-party system170. As an example and not by way of limitation, a profile interface corresponding to a particular external web interface may be the particular external web interface and the profile interface may correspond to aparticular concept node1004. Profile interfaces may be viewable by all or a selected subset of other users. As an example and not by way of limitation, auser node1002 may have a corresponding user-profile interface in which the corresponding user may add content, make declarations, or otherwise express himself or herself. As another example and not by way of limitation, aconcept node1004 may have a corresponding concept-profile interface in which one or more users may add content, make declarations, or express themselves, particularly in relation to the concept corresponding toconcept node1004.
• In particular embodiments, aconcept node1004 may represent a third-party web interface or resource hosted by a third-party system170. The third-party web interface or resource may include, among other elements, content, a selectable or other icon, or other inter-actable object representing an action or activity. As an example and not by way of limitation, a third-party web interface may include a selectable icon such as “like,” “check-in,” “eat,” “recommend,” or another suitable action or activity. A user viewing the third-party web interface may perform an action by selecting one of the icons (e.g., “check-in”), causing aclient system130 to send to the social-networking system160 a message indicating the user's action. In response to the message, the social-networking system160 may create an edge (e.g., a check-in-type edge) between auser node1002 corresponding to the user and aconcept node1004 corresponding to the third-party web interface or resource andstore edge1006 in one or more data stores.
• In particular embodiments, a pair of nodes in thesocial graph1000 may be connected to each other by one ormore edges1006. Anedge1006 connecting a pair of nodes may represent a relationship between the pair of nodes. In particular embodiments, anedge1006 may include or represent one or more data objects or attributes corresponding to the relationship between a pair of nodes. As an example and not by way of limitation, a first user may indicate that a second user is a “friend” of the first user. In response to this indication, the social-networking system160 may send a “friend request” to the second user. If the second user confirms the “friend request,” the social-networking system160 may create anedge1006 connecting the first user'suser node1002 to the second user'suser node1002 in thesocial graph1000 andstore edge1006 as social-graph information in one or more ofdata stores164. In the example ofFIG. 10, thesocial graph1000 includes anedge1006 indicating a friend relation betweenuser nodes1002 of user “A” and user “B” and an edge indicating a friend relation betweenuser nodes1002 of user “C” and user “B.” Although this disclosure describes or illustratesparticular edges1006 with particular attributes connectingparticular user nodes1002, this disclosure contemplates anysuitable edges1006 with any suitable attributes connectinguser nodes1002. As an example and not by way of limitation, anedge1006 may represent a friendship, family relationship, business or employment relationship, fan relationship (including, e.g., liking, etc.), follower relationship, visitor relationship (including, e.g., accessing, viewing, checking-in, sharing, etc.), subscriber relationship, superior/subordinate relationship, reciprocal relationship, non-reciprocal relationship, another suitable type of relationship, or two or more such relationships. Moreover, although this disclosure generally describes nodes as being connected, this disclosure also describes users or concepts as being connected. Herein, references to users or concepts being connected may, where appropriate, refer to the nodes corresponding to those users or concepts being connected in thesocial graph1000 by one ormore edges1006.
• In particular embodiments, anedge1006 between auser node1002 and aconcept node1004 may represent a particular action or activity performed by a user associated withuser node1002 toward a concept associated with aconcept node1004. As an example and not by way of limitation, as illustrated inFIG. 10, a user may “like,” “attended,” “played,” “listened,” “cooked,” “worked at,” or “watched” a concept, each of which may correspond to an edge type or subtype. A concept-profile interface corresponding to aconcept node1004 may include, for example, a selectable “check in” icon (such as, for example, a clickable “check in” icon) or a selectable “add to favorites” icon. Similarly, after a user clicks these icons, the social-networking system160 may create a “favorite” edge or a “check in” edge in response to a user's action corresponding to a respective action. As another example and not by way of limitation, a user (user “C”) may listen to a particular song (“Song Name”) using a particular application (App “Online Music”, which is an online music application). In this case, the social-networking system160 may create a “listened”edge1006 and a “used” edge (as illustrated inFIG. 10) betweenuser nodes1002 corresponding to the user andconcept nodes1004 corresponding to the song and application to indicate that the user listened to the song and used the application. Moreover, the social-networking system160 may create a “played” edge1006 (as illustrated inFIG. 10) betweenconcept nodes1004 corresponding to the song and the application to indicate that the particular song was played by the particular application. In this case, “played”edge1006 corresponds to an action performed by an external application (App “Online Music”) on an external audio file (the song “Song Name”). Although this disclosure describesparticular edges1006 with particular attributes connectinguser nodes1002 andconcept nodes1004, this disclosure contemplates anysuitable edges1006 with any suitable attributes connectinguser nodes1002 andconcept nodes1004. Moreover, although this disclosure describes edges between auser node1002 and aconcept node1004 representing a single relationship, this disclosure contemplates edges between auser node1002 and aconcept node1004 representing one or more relationships. As an example and not by way of limitation, anedge1006 may represent both that a user likes and has used at a particular concept. Alternatively, anotheredge1006 may represent each type of relationship (or multiples of a single relationship) between auser node1002 and a concept node1004 (as illustrated inFIG. 10 betweenuser node1002 for user “E” andconcept node1004 for “App ‘Online Music’”).
• In particular embodiments, the social-networking system160 may create anedge1006 between auser node1002 and aconcept node1004 in thesocial graph1000. As an example and not by way of limitation, a user viewing a concept-profile interface (such as, for example, by using a web browser or a special-purpose application hosted by the user's client system130) may indicate that he or she likes the concept represented by theconcept node1004 by clicking or selecting a “Like” icon, which may cause the user'sclient system130 to send to the social-networking system160 a message indicating the user's liking of the concept associated with the concept-profile interface. In response to the message, the social-networking system160 may create anedge1006 betweenuser node1002 associated with the user andconcept node1004, as illustrated by “like”edge1006 between the user andconcept node1004. In particular embodiments, the social-networking system160 may store anedge1006 in one or more data stores. In particular embodiments, anedge1006 may be automatically formed by the social-networking system160 in response to a particular user action. As an example and not by way of limitation, if a first user uploads a picture, watches a movie, or listens to a song, anedge1006 may be formed betweenuser node1002 corresponding to the first user andconcept nodes1004 corresponding to those concepts. Although this disclosure describes formingparticular edges1006 in particular manners, this disclosure contemplates forming anysuitable edges1006 in any suitable manner.
Vector Spaces and Embeddings
• FIG. 11 illustrates an example view of avector space1100. In particular embodiments, an object or an n-gram may be represented in a d-dimensional vector space, where d denotes any suitable number of dimensions. Although thevector space1100 is illustrated as a three-dimensional space, this is for illustrative purposes only, as thevector space1100 may be of any suitable dimension. In particular embodiments, an n-gram may be represented in thevector space1100 as a vector referred to as a term embedding. Each vector may comprise coordinates corresponding to a particular point in the vector space1100 (i.e., the terminal point of the vector). As an example and not by way of limitation,vectors1110,1120, and1130 may be represented as points in thevector space1100, as illustrated inFIG. 11. An n-gram may be mapped to a respective vector representation. As an example and not by way of limitation, n-grams t₁and t₂may be mapped to vectors

  

  and

  

  in thevector space1100, respectively, by applying a function

  

  defined by a dictionary, such that

  

  =

  

  (t₁) and

  

  =

  

  (t₂). As another example and not by way of limitation, a dictionary trained to map text to a vector representation may be utilized, or such a dictionary may be itself generated via training. As another example and not by way of limitation, a model, such as Word2vec, may be used to map an n-gram to a vector representation in thevector space1100. In particular embodiments, an n-gram may be mapped to a vector representation in thevector space1100 by using a machine leaning model (e.g., a neural network). The machine learning model may have been trained using a sequence of training data (e.g., a corpus of objects each comprising n-grams).
• In particular embodiments, an object may be represented in thevector space1100 as a vector referred to as a feature vector or an object embedding. As an example and not by way of limitation, objects e₁and e₂may be mapped to vectors

  

  and

  

  in thevector space1100, respectively, by applying a function

  

  , such that

  

  =

  

  (e₁) and

  

  =

  

  (e₂). In particular embodiments, an object may be mapped to a vector based on one or more properties, attributes, or features of the object, relationships of the object with other objects, or any other suitable information associated with the object. As an example and not by way of limitation, a function

  

  may map objects to vectors by feature extraction, which may start from an initial set of measured data and build derived values (e.g., features). As an example and not by way of limitation, an object comprising a video or an image may be mapped to a vector by using an algorithm to detect or isolate various desired portions or shapes of the object. Features used to calculate the vector may be based on information obtained from edge detection, corner detection, blob detection, ridge detection, scale-invariant feature transformation, edge direction, changing intensity, autocorrelation, motion detection, optical flow, thresholding, blob extraction, template matching, Hough transformation (e.g., lines, circles, ellipses, arbitrary shapes), or any other suitable information. As another example and not by way of limitation, an object comprising audio data may be mapped to a vector based on features such as a spectral slope, a tonality coefficient, an audio spectrum centroid, an audio spectrum envelope, a Mel-frequency cepstrum, or any other suitable information. In particular embodiments, when an object has data that is either too large to be efficiently processed or comprises redundant data, a function

  

  may map the object to a vector using a transformed reduced set of features (e.g., feature selection). In particular embodiments, a function

  

  may map an object e to a vector

  

  (e) based on one or more n-grams associated with object e. Although this disclosure describes representing an n-gram or an object in a vector space in a particular manner, this disclosure contemplates representing an n-gram or an object in a vector space in any suitable manner.
• In particular embodiments, the social-networking system160 may calculate a similarity metric of vectors invector space1100. A similarity metric may be a cosine similarity, a Minkowski distance, a Mahalanobis distance, a Jaccard similarity coefficient, or any suitable similarity metric. As an example and not by way of limitation, a similarity metric of

  

  and

  

  may be a cosine similarity
• $\frac{\overrightarrow{v_1}·\overrightarrow{v_2}}{\overrightarrow{v_1}\overrightarrow{v_2}}.$
• As another example and not by way of limitation, a similarity metric of

  

  and

  

  may be a Euclidean distance ∥

  

  −

  

  ∥. A similarity metric of two vectors may represent how similar the two objects or n-grams corresponding to the two vectors, respectively, are to one another, as measured by the distance between the two vectors in thevector space1100. As an example and not by way of limitation,vector1110 andvector1120 may correspond to objects that are more similar to one another than the objects corresponding tovector1110 andvector1130, based on the distance between the respective vectors. Although this disclosure describes calculating a similarity metric between vectors in a particular manner, this disclosure contemplates calculating a similarity metric between vectors in any suitable manner.
• More information on vector spaces, embeddings, feature vectors, and similarity metrics may be found in U.S. patent application Ser. No. 14/949,436, filed 23 Nov. 2015, U.S. patent application Ser. No. 15/286,315, filed 5 Oct. 2016, and U.S. patent application Ser. No. 15/365,789, filed 30 Nov. 2016, each of which is incorporated by reference.
Artificial Neural Networks
• FIG. 12 illustrates an example artificial neural network (“ANN”)1200. In particular embodiments, an ANN may refer to a computational model comprising one or more nodes.Example ANN1200 may comprise aninput layer1210, hiddenlayers1220,1230,1240, and anoutput layer1250. Each layer of theANN1200 may comprise one or more nodes, such as anode1205 or anode1215. In particular embodiments, each node of an ANN may be connected to another node of the ANN. As an example and not by way of limitation, each node of theinput layer1210 may be connected to one of more nodes of the hiddenlayer1220. In particular embodiments, one or more nodes may be a bias node (e.g., a node in a layer that is not connected to and does not receive input from any node in a previous layer). In particular embodiments, each node in each layer may be connected to one or more nodes of a previous or subsequent layer. AlthoughFIG. 12 depicts a particular ANN with a particular number of layers, a particular number of nodes, and particular connections between nodes, this disclosure contemplates any suitable ANN with any suitable number of layers, any suitable number of nodes, and any suitable connections between nodes. As an example and not by way of limitation, althoughFIG. 12 depicts a connection between each node of theinput layer1210 and each node of the hiddenlayer1220, one or more nodes of theinput layer1210 may not be connected to one or more nodes of the hiddenlayer1220.
• In particular embodiments, an ANN may be a feedforward ANN (e.g., an ANN with no cycles or loops where communication between nodes flows in one direction beginning with the input layer and proceeding to successive layers). As an example and not by way of limitation, the input to each node of the hiddenlayer1220 may comprise the output of one or more nodes of theinput layer1210. As another example and not by way of limitation, the input to each node of theoutput layer1250 may comprise the output of one or more nodes of the hiddenlayer1240. In particular embodiments, an ANN may be a deep neural network (e.g., a neural network comprising at least two hidden layers). In particular embodiments, an ANN may be a deep residual network. A deep residual network may be a feedforward ANN comprising hidden layers organized into residual blocks. The input into each residual block after the first residual block may be a function of the output of the previous residual block and the input of the previous residual block. As an example and not by way of limitation, the input into residual block N may be F(x)+x, where F(x) may be the output of residual block N−1, x may be the input into residual block N−1. Although this disclosure describes a particular ANN, this disclosure contemplates any suitable ANN.
• In particular embodiments, an activation function may correspond to each node of an ANN. An activation function of a node may define the output of a node for a given input. In particular embodiments, an input to a node may comprise a set of inputs. As an example and not by way of limitation, an activation function may be an identity function, a binary step function, a logistic function, or any other suitable function. As another example and not by way of limitation, an activation function for a node k may be the sigmoid function
• $F_k\left(s_k\right)=\frac{1}{1+e^{-s_k}},$
• the hyperbolic tangent function
• $F_k\left(s_k\right)=\frac{e^{s_k}-e^{-s_k}}{e^{s_k}+e^{-s_k}},$
• the rectifier F_k(S_k)=max (0, s_k), or any other suitable function F_k(s_k), where s_kmay be the effective input to node k. In particular embodiments, the input of an activation function corresponding to a node may be weighted. Each node may generate output using a corresponding activation function based on weighted inputs. In particular embodiments, each connection between nodes may be associated with a weight. As an example and not by way of limitation, aconnection1225 between thenode1205 and thenode1215 may have a weighting coefficient of 0.4, which may indicate that 0.4 multiplied by the output of thenode1205 is used as an input to thenode1215. As another example and not by way of limitation, the output y_kof node k may be y_k=F_k(S_k), where F_kmay be the activation function corresponding to node k, s_k=Σ_j(w_jkx_j) may be the effective input to node k, x_jmay be the output of a node j connected to node k, and w_jkmay be the weighting coefficient between node j and node k. In particular embodiments, the input to nodes of the input layer may be based on a vector representing an object. Although this disclosure describes particular inputs to and outputs of nodes, this disclosure contemplates any suitable inputs to and outputs of nodes. Moreover, although this disclosure may describe particular connections and weights between nodes, this disclosure contemplates any suitable connections and weights between nodes.
• In particular embodiments, an ANN may be trained using training data. As an example and not by way of limitation, training data may comprise inputs to theANN1200 and an expected output. As another example and not by way of limitation, training data may comprise vectors each representing a training object and an expected label for each training object. In particular embodiments, training an ANN may comprise modifying the weights associated with the connections between nodes of the ANN by optimizing an objective function. As an example and not by way of limitation, a training method may be used (e.g., the conjugate gradient method, the gradient descent method, the stochastic gradient descent) to backpropagate the sum-of-squares error measured as a distances between each vector representing a training object (e.g., using a cost function that minimizes the sum-of-squares error). In particular embodiments, an ANN may be trained using a dropout technique. As an example and not by way of limitation, one or more nodes may be temporarily omitted (e.g., receive no input and generate no output) while training. For each training object, one or more nodes of the ANN may have some probability of being omitted. The nodes that are omitted for a particular training object may be different than the nodes omitted for other training objects (e.g., the nodes may be temporarily omitted on an object-by-object basis). Although this disclosure describes training an ANN in a particular manner, this disclosure contemplates training an ANN in any suitable manner.
Privacy
• In particular embodiments, one or more objects (e.g., content or other types of objects) of a computing system may be associated with one or more privacy settings. The one or more objects may be stored on or otherwise associated with any suitable computing system or application, such as, for example, a social-networking system160, aclient system130, anassistant system140, a third-party system170, a social-networking application, an assistant application, a messaging application, a photo-sharing application, or any other suitable computing system or application. Although the examples discussed herein are in the context of an online social network, these privacy settings may be applied to any other suitable computing system. Privacy settings (or “access settings”) for an object may be stored in any suitable manner, such as, for example, in association with the object, in an index on an authorization server, in another suitable manner, or any suitable combination thereof. A privacy setting for an object may specify how the object (or particular information associated with the object) can be accessed, stored, or otherwise used (e.g., viewed, shared, modified, copied, executed, surfaced, or identified) within the online social network. When privacy settings for an object allow a particular user or other entity to access that object, the object may be described as being “visible” with respect to that user or other entity. As an example and not by way of limitation, a user of the online social network may specify privacy settings for a user-profile page that identify a set of users that may access work-experience information on the user-profile page, thus excluding other users from accessing that information.
• In particular embodiments, privacy settings for an object may specify a “blocked list” of users or other entities that should not be allowed to access certain information associated with the object. In particular embodiments, the blocked list may include third-party entities. The blocked list may specify one or more users or entities for which an object is not visible. As an example and not by way of limitation, a user may specify a set of users who may not access photo albums associated with the user, thus excluding those users from accessing the photo albums (while also possibly allowing certain users not within the specified set of users to access the photo albums). In particular embodiments, privacy settings may be associated with particular social-graph elements. Privacy settings of a social-graph element, such as a node or an edge, may specify how the social-graph element, information associated with the social-graph element, or objects associated with the social-graph element can be accessed using the online social network. As an example and not by way of limitation, aparticular concept node1004 corresponding to a particular photo may have a privacy setting specifying that the photo may be accessed only by users tagged in the photo and friends of the users tagged in the photo. In particular embodiments, privacy settings may allow users to opt in to or opt out of having their content, information, or actions stored/logged by the social-networking system160 orassistant system140 or shared with other systems (e.g., a third-party system170). Although this disclosure describes using particular privacy settings in a particular manner, this disclosure contemplates using any suitable privacy settings in any suitable manner.
• In particular embodiments, privacy settings may be based on one or more nodes or edges of asocial graph1000. A privacy setting may be specified for one ormore edges1006 or edge-types of thesocial graph1000, or with respect to one ormore nodes1002,1004 or node-types of thesocial graph1000. The privacy settings applied to aparticular edge1006 connecting two nodes may control whether the relationship between the two entities corresponding to the nodes is visible to other users of the online social network. Similarly, the privacy settings applied to a particular node may control whether the user or concept corresponding to the node is visible to other users of the online social network. As an example and not by way of limitation, a first user may share an object to the social-networking system160. The object may be associated with aconcept node1004 connected to auser node1002 of the first user by anedge1006. The first user may specify privacy settings that apply to aparticular edge1006 connecting to theconcept node1004 of the object, or may specify privacy settings that apply to alledges1006 connecting to theconcept node1004. As another example and not by way of limitation, the first user may share a set of objects of a particular object-type (e.g., a set of images). The first user may specify privacy settings with respect to all objects associated with the first user of that particular object-type as having a particular privacy setting (e.g., specifying that all images posted by the first user are visible only to friends of the first user and/or users tagged in the images).
• In particular embodiments, the social-networking system160 may present a “privacy wizard” (e.g., within a webpage, a module, one or more dialog boxes, or any other suitable interface) to the first user to assist the first user in specifying one or more privacy settings. The privacy wizard may display instructions, suitable privacy-related information, current privacy settings, one or more input fields for accepting one or more inputs from the first user specifying a change or confirmation of privacy settings, or any suitable combination thereof. In particular embodiments, the social-networking system160 may offer a “dashboard” functionality to the first user that may display, to the first user, current privacy settings of the first user. The dashboard functionality may be displayed to the first user at any appropriate time (e.g., following an input from the first user summoning the dashboard functionality, following the occurrence of a particular event or trigger action). The dashboard functionality may allow the first user to modify one or more of the first user's current privacy settings at any time, in any suitable manner (e.g., redirecting the first user to the privacy wizard).
• Privacy settings associated with an object may specify any suitable granularity of permitted access or denial of access. As an example and not by way of limitation, access or denial of access may be specified for particular users (e.g., only me, my roommates, my boss), users within a particular degree-of-separation (e.g., friends, friends-of-friends), user groups (e.g., the gaming club, my family), user networks (e.g., employees of particular employers, students or alumni of particular university), all users (“public”), no users (“private”), users of third-party systems170, particular applications (e.g., third-party applications, external websites), other suitable entities, or any suitable combination thereof. Although this disclosure describes particular granularities of permitted access or denial of access, this disclosure contemplates any suitable granularities of permitted access or denial of access.
• In particular embodiments, one ormore servers162 may be authorization/privacy servers for enforcing privacy settings. In response to a request from a user (or other entity) for a particular object stored in adata store164, the social-networking system160 may send a request to thedata store164 for the object. The request may identify the user associated with the request and the object may be sent only to the user (or aclient system130 of the user) if the authorization server determines that the user is authorized to access the object based on the privacy settings associated with the object. If the requesting user is not authorized to access the object, the authorization server may prevent the requested object from being retrieved from thedata store164 or may prevent the requested object from being sent to the user. In the search-query context, an object may be provided as a search result only if the querying user is authorized to access the object, e.g., if the privacy settings for the object allow it to be surfaced to, discovered by, or otherwise visible to the querying user. In particular embodiments, an object may represent content that is visible to a user through a newsfeed of the user. As an example and not by way of limitation, one or more objects may be visible to a user's “Trending” page. In particular embodiments, an object may correspond to a particular user. The object may be content associated with the particular user, or may be the particular user's account or information stored on the social-networking system160, or other computing system. As an example and not by way of limitation, a first user may view one or more second users of an online social network through a “People You May Know” function of the online social network, or by viewing a list of friends of the first user. As an example and not by way of limitation, a first user may specify that they do not wish to see objects associated with a particular second user in their newsfeed or friends list. If the privacy settings for the object do not allow it to be surfaced to, discovered by, or visible to the user, the object may be excluded from the search results. Although this disclosure describes enforcing privacy settings in a particular manner, this disclosure contemplates enforcing privacy settings in any suitable manner.
• In particular embodiments, different objects of the same type associated with a user may have different privacy settings. Different types of objects associated with a user may have different types of privacy settings. As an example and not by way of limitation, a first user may specify that the first user's status updates are public, but any images shared by the first user are visible only to the first user's friends on the online social network. As another example and not by way of limitation, a user may specify different privacy settings for different types of entities, such as individual users, friends-of-friends, followers, user groups, or corporate entities. As another example and not by way of limitation, a first user may specify a group of users that may view videos posted by the first user, while keeping the videos from being visible to the first user's employer. In particular embodiments, different privacy settings may be provided for different user groups or user demographics. As an example and not by way of limitation, a first user may specify that other users who attend the same university as the first user may view the first user's pictures, but that other users who are family members of the first user may not view those same pictures.
• In particular embodiments, the social-networking system160 may provide one or more default privacy settings for each object of a particular object-type. A privacy setting for an object that is set to a default may be changed by a user associated with that object. As an example and not by way of limitation, all images posted by a first user may have a default privacy setting of being visible only to friends of the first user and, for a particular image, the first user may change the privacy setting for the image to be visible to friends and friends-of-friends.
• In particular embodiments, privacy settings may allow a first user to specify (e.g., by opting out, by not opting in) whether the social-networking system160 orassistant system140 may receive, collect, log, or store particular objects or information associated with the user for any purpose. In particular embodiments, privacy settings may allow the first user to specify whether particular applications or processes may access, store, or use particular objects or information associated with the user. The privacy settings may allow the first user to opt in or opt out of having objects or information accessed, stored, or used by specific applications or processes. The social-networking system160 orassistant system140 may access such information in order to provide a particular function or service to the first user, without the social-networking system160 orassistant system140 having access to that information for any other purposes. Before accessing, storing, or using such objects or information, the social-networking system160 orassistant system140 may prompt the user to provide privacy settings specifying which applications or processes, if any, may access, store, or use the object or information prior to allowing any such action. As an example and not by way of limitation, a first user may transmit a message to a second user via an application related to the online social network (e.g., a messaging app), and may specify privacy settings that such messages should not be stored by the social-networking system160 orassistant system140.
• In particular embodiments, a user may specify whether particular types of objects or information associated with the first user may be accessed, stored, or used by the social-networking system160 orassistant system140. As an example and not by way of limitation, the first user may specify that images sent by the first user through the social-networking system160 orassistant system140 may not be stored by the social-networking system160 orassistant system140. As another example and not by way of limitation, a first user may specify that messages sent from the first user to a particular second user may not be stored by the social-networking system160 orassistant system140. As yet another example and not by way of limitation, a first user may specify that all objects sent via a particular application may be saved by the social-networking system160 orassistant system140.
• In particular embodiments, privacy settings may allow a first user to specify whether particular objects or information associated with the first user may be accessed fromparticular client systems130 or third-party systems170. The privacy settings may allow the first user to opt in or opt out of having objects or information accessed from a particular device (e.g., the phone book on a user's smart phone), from a particular application (e.g., a messaging app), or from a particular system (e.g., an email server). The social-networking system160 orassistant system140 may provide default privacy settings with respect to each device, system, or application, and/or the first user may be prompted to specify a particular privacy setting for each context. As an example and not by way of limitation, the first user may utilize a location-services feature of the social-networking system160 orassistant system140 to provide recommendations for restaurants or other places in proximity to the user. The first user's default privacy settings may specify that the social-networking system160 orassistant system140 may use location information provided from aclient device130 of the first user to provide the location-based services, but that the social-networking system160 orassistant system140 may not store the location information of the first user or provide it to any third-party system170. The first user may then update the privacy settings to allow location information to be used by a third-party image-sharing application in order to geo-tag photos.
• In particular embodiments, privacy settings may allow a user to specify one or more geographic locations from which objects can be accessed. Access or denial of access to the objects may depend on the geographic location of a user who is attempting to access the objects. As an example and not by way of limitation, a user may share an object and specify that only users in the same city may access or view the object. As another example and not by way of limitation, a first user may share an object and specify that the object is visible to second users only while the first user is in a particular location. If the first user leaves the particular location, the object may no longer be visible to the second users. As another example and not by way of limitation, a first user may specify that an object is visible only to second users within a threshold distance from the first user. If the first user subsequently changes location, the original second users with access to the object may lose access, while a new group of second users may gain access as they come within the threshold distance of the first user.
• In particular embodiments, the social-networking system160 orassistant system140 may have functionalities that may use, as inputs, personal or biometric information of a user for user-authentication or experience-personalization purposes. A user may opt to make use of these functionalities to enhance their experience on the online social network. As an example and not by way of limitation, a user may provide personal or biometric information to the social-networking system160 orassistant system140. The user's privacy settings may specify that such information may be used only for particular processes, such as authentication, and further specify that such information may not be shared with any third-party system170 or used for other processes or applications associated with the social-networking system160 orassistant system140. As another example and not by way of limitation, the social-networking system160 may provide a functionality for a user to provide voice-print recordings to the online social network. As an example and not by way of limitation, if a user wishes to utilize this function of the online social network, the user may provide a voice recording of his or her own voice to provide a status update on the online social network. The recording of the voice-input may be compared to a voice print of the user to determine what words were spoken by the user. The user's privacy setting may specify that such voice recording may be used only for voice-input purposes (e.g., to authenticate the user, to send voice messages, to improve voice recognition in order to use voice-operated features of the online social network), and further specify that such voice recording may not be shared with any third-party system170 or used by other processes or applications associated with the social-networking system160. As another example and not by way of limitation, the social-networking system160 may provide a functionality for a user to provide a reference image (e.g., a facial profile, a retinal scan) to the online social network. The online social network may compare the reference image against a later-received image input (e.g., to authenticate the user, to tag the user in photos). The user's privacy setting may specify that such image may be used only for a limited purpose (e.g., authentication, tagging the user in photos), and further specify that such image may not be shared with any third-party system170 or used by other processes or applications associated with the social-networking system160.
Systems and Methods
• FIG. 13 illustrates anexample computer system1300. In particular embodiments, one ormore computer systems1300 perform one or more steps of one or more methods described or illustrated herein. In particular embodiments, one ormore computer systems1300 provide functionality described or illustrated herein. In particular embodiments, software running on one ormore computer systems1300 performs one or more steps of one or more methods described or illustrated herein or provides functionality described or illustrated herein. Particular embodiments include one or more portions of one ormore computer systems1300. Herein, reference to a computer system may encompass a computing device, and vice versa, where appropriate. Moreover, reference to a computer system may encompass one or more computer systems, where appropriate.
• This disclosure contemplates any suitable number ofcomputer systems1300. This disclosure contemplatescomputer system1300 taking any suitable physical form. As example and not by way of limitation,computer system1300 may be an embedded computer system, a system-on-chip (SOC), a single-board computer system (SBC) (such as, for example, a computer-on-module (COM) or system-on-module (SOM)), a desktop computer system, a laptop or notebook computer system, an interactive kiosk, a mainframe, a mesh of computer systems, a mobile telephone, a personal digital assistant (PDA), a server, a tablet computer system, or a combination of two or more of these. Where appropriate,computer system1300 may include one ormore computer systems1300; be unitary or distributed; span multiple locations; span multiple machines; span multiple data centers; or reside in a cloud, which may include one or more cloud components in one or more networks. Where appropriate, one ormore computer systems1300 may perform without substantial spatial or temporal limitation one or more steps of one or more methods described or illustrated herein. As an example and not by way of limitation, one ormore computer systems1300 may perform in real time or in batch mode one or more steps of one or more methods described or illustrated herein. One ormore computer systems1300 may perform at different times or at different locations one or more steps of one or more methods described or illustrated herein, where appropriate.
• In particular embodiments,computer system1300 includes aprocessor1302,memory1304,storage1306, an input/output (I/O)interface1308, acommunication interface1310, and abus1312. Although this disclosure describes and illustrates a particular computer system having a particular number of particular components in a particular arrangement, this disclosure contemplates any suitable computer system having any suitable number of any suitable components in any suitable arrangement.
• In particular embodiments,processor1302 includes hardware for executing instructions, such as those making up a computer program. As an example and not by way of limitation, to execute instructions,processor1302 may retrieve (or fetch) the instructions from an internal register, an internal cache,memory1304, orstorage1306; decode and execute them; and then write one or more results to an internal register, an internal cache,memory1304, orstorage1306. In particular embodiments,processor1302 may include one or more internal caches for data, instructions, or addresses. This disclosure contemplatesprocessor1302 including any suitable number of any suitable internal caches, where appropriate. As an example and not by way of limitation,processor1302 may include one or more instruction caches, one or more data caches, and one or more translation lookaside buffers (TLBs). Instructions in the instruction caches may be copies of instructions inmemory1304 orstorage1306, and the instruction caches may speed up retrieval of those instructions byprocessor1302. Data in the data caches may be copies of data inmemory1304 orstorage1306 for instructions executing atprocessor1302 to operate on; the results of previous instructions executed atprocessor1302 for access by subsequent instructions executing atprocessor1302 or for writing tomemory1304 orstorage1306; or other suitable data. The data caches may speed up read or write operations byprocessor1302. The TLBs may speed up virtual-address translation forprocessor1302. In particular embodiments,processor1302 may include one or more internal registers for data, instructions, or addresses. This disclosure contemplatesprocessor1302 including any suitable number of any suitable internal registers, where appropriate. Where appropriate,processor1302 may include one or more arithmetic logic units (ALUs); be a multi-core processor; or include one ormore processors1302. Although this disclosure describes and illustrates a particular processor, this disclosure contemplates any suitable processor.
• In particular embodiments,memory1304 includes main memory for storing instructions forprocessor1302 to execute or data forprocessor1302 to operate on. As an example and not by way of limitation,computer system1300 may load instructions fromstorage1306 or another source (such as, for example, another computer system1300) tomemory1304.Processor1302 may then load the instructions frommemory1304 to an internal register or internal cache. To execute the instructions,processor1302 may retrieve the instructions from the internal register or internal cache and decode them. During or after execution of the instructions,processor1302 may write one or more results (which may be intermediate or final results) to the internal register or internal cache.Processor1302 may then write one or more of those results tomemory1304. In particular embodiments,processor1302 executes only instructions in one or more internal registers or internal caches or in memory1304 (as opposed tostorage1306 or elsewhere) and operates only on data in one or more internal registers or internal caches or in memory1304 (as opposed tostorage1306 or elsewhere). One or more memory buses (which may each include an address bus and a data bus) may coupleprocessor1302 tomemory1304.Bus1312 may include one or more memory buses, as described below. In particular embodiments, one or more memory management units (MMUs) reside betweenprocessor1302 andmemory1304 and facilitate accesses tomemory1304 requested byprocessor1302. In particular embodiments,memory1304 includes random access memory (RAM). This RAM may be volatile memory, where appropriate. Where appropriate, this RAM may be dynamic RAM (DRAM) or static RAM (SRAM). Moreover, where appropriate, this RAM may be single-ported or multi-ported RAM. This disclosure contemplates any suitable RAM.Memory1304 may include one ormore memories1304, where appropriate. Although this disclosure describes and illustrates particular memory, this disclosure contemplates any suitable memory.
• In particular embodiments,storage1306 includes mass storage for data or instructions. As an example and not by way of limitation,storage1306 may include a hard disk drive (HDD), a floppy disk drive, flash memory, an optical disc, a magneto-optical disc, magnetic tape, or a Universal Serial Bus (USB) drive or a combination of two or more of these.Storage1306 may include removable or non-removable (or fixed) media, where appropriate.Storage1306 may be internal or external tocomputer system1300, where appropriate. In particular embodiments,storage1306 is non-volatile, solid-state memory. In particular embodiments,storage1306 includes read-only memory (ROM). Where appropriate, this ROM may be mask-programmed ROM, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), electrically alterable ROM (EAROM), or flash memory or a combination of two or more of these. This disclosure contemplatesmass storage1306 taking any suitable physical form.Storage1306 may include one or more storage control units facilitating communication betweenprocessor1302 andstorage1306, where appropriate. Where appropriate,storage1306 may include one ormore storages1306. Although this disclosure describes and illustrates particular storage, this disclosure contemplates any suitable storage.
• In particular embodiments, I/O interface1308 includes hardware, software, or both, providing one or more interfaces for communication betweencomputer system1300 and one or more I/O devices.Computer system1300 may include one or more of these I/O devices, where appropriate. One or more of these I/O devices may enable communication between a person andcomputer system1300. As an example and not by way of limitation, an I/O device may include a keyboard, keypad, microphone, monitor, mouse, printer, scanner, speaker, still camera, stylus, tablet, touch screen, trackball, video camera, another suitable I/O device or a combination of two or more of these. An I/O device may include one or more sensors. This disclosure contemplates any suitable I/O devices and any suitable I/O interfaces1308 for them. Where appropriate, I/O interface1308 may include one or more device or softwaredrivers enabling processor1302 to drive one or more of these I/O devices. I/O interface1308 may include one or more I/O interfaces1308, where appropriate. Although this disclosure describes and illustrates a particular I/O interface, this disclosure contemplates any suitable I/O interface.
• In particular embodiments,communication interface1310 includes hardware, software, or both providing one or more interfaces for communication (such as, for example, packet-based communication) betweencomputer system1300 and one or moreother computer systems1300 or one or more networks. As an example and not by way of limitation,communication interface1310 may include a network interface controller (NIC) or network adapter for communicating with an Ethernet or other wire-based network or a wireless NIC (WNIC) or wireless adapter for communicating with a wireless network, such as a WI-FI network. This disclosure contemplates any suitable network and anysuitable communication interface1310 for it. As an example and not by way of limitation,computer system1300 may communicate with an ad hoc network, a personal area network (PAN), a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), or one or more portions of the Internet or a combination of two or more of these. One or more portions of one or more of these networks may be wired or wireless. As an example,computer system1300 may communicate with a wireless PAN (WPAN) (such as, for example, a BLUETOOTH WPAN), a WI-FI network, a WI-MAX network, a cellular telephone network (such as, for example, a Global System for Mobile Communications (GSM) network), or other suitable wireless network or a combination of two or more of these.Computer system1300 may include anysuitable communication interface1310 for any of these networks, where appropriate.Communication interface1310 may include one ormore communication interfaces1310, where appropriate. Although this disclosure describes and illustrates a particular communication interface, this disclosure contemplates any suitable communication interface.
• In particular embodiments,bus1312 includes hardware, software, or both coupling components ofcomputer system1300 to each other. As an example and not by way of limitation,bus1312 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a front-side bus (FSB), a HYPERTRANSPORT (HT) interconnect, an Industry Standard Architecture (ISA) bus, an INFINIBAND interconnect, a low-pin-count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCIe) bus, a serial advanced technology attachment (SATA) bus, a Video Electronics Standards Association local (VLB) bus, or another suitable bus or a combination of two or more of these.Bus1312 may include one ormore buses1312, where appropriate. Although this disclosure describes and illustrates a particular bus, this disclosure contemplates any suitable bus or interconnect.
• Herein, a computer-readable non-transitory storage medium or media may include one or more semiconductor-based or other integrated circuits (ICs) (such, as for example, field-programmable gate arrays (FPGAs) or application-specific ICs (ASICs)), hard disk drives (HDDs), hybrid hard drives (HHDs), optical discs, optical disc drives (ODDs), magneto-optical discs, magneto-optical drives, floppy diskettes, floppy disk drives (FDDs), magnetic tapes, solid-state drives (SSDs), RAM-drives, SECURE DIGITAL cards or drives, any other suitable computer-readable non-transitory storage media, or any suitable combination of two or more of these, where appropriate. A computer-readable non-transitory storage medium may be volatile, non-volatile, or a combination of volatile and non-volatile, where appropriate.
Miscellaneous
• Herein, “or” is inclusive and not exclusive, unless expressly indicated otherwise or indicated otherwise by context. Therefore, herein, “A or B” means “A, B, or both,” unless expressly indicated otherwise or indicated otherwise by context. Moreover, “and” is both joint and several, unless expressly indicated otherwise or indicated otherwise by context. Therefore, herein, “A and B” means “A and B, jointly or severally,” unless expressly indicated otherwise or indicated otherwise by context.
• The scope of this disclosure encompasses all changes, substitutions, variations, alterations, and modifications to the example embodiments described or illustrated herein that a person having ordinary skill in the art would comprehend. The scope of this disclosure is not limited to the example embodiments described or illustrated herein. Moreover, although this disclosure describes and illustrates respective embodiments herein as including particular components, elements, feature, functions, operations, or steps, any of these embodiments may include any combination or permutation of any of the components, elements, features, functions, operations, or steps described or illustrated anywhere herein that a person having ordinary skill in the art would comprehend. Furthermore, reference in the appended claims to an apparatus or system or a component of an apparatus or system being adapted to, arranged to, capable of, configured to, enabled to, operable to, or operative to perform a particular function encompasses that apparatus, system, component, whether or not it or that particular function is activated, turned on, or unlocked, as long as that apparatus, system, or component is so adapted, arranged, capable, configured, enabled, operable, or operative. Additionally, although this disclosure describes or illustrates particular embodiments as providing particular advantages, particular embodiments may provide none, some, or all of these advantages.

Claims (20)

What is claimed is:

1. A method comprising, by a client system:

receiving, at the client system, an initial input from a first user, wherein the initial input is in a first modality;

determining one or more intents and one or more slots corresponding to the initial input, wherein the one or more slots are conditioned on the one or more intents;

generating, based on the one or more intents and the one or more slots, one or more candidate continuation-inputs, where the one or more candidate continuation-inputs are in one or more candidate modalities, respectively, wherein the candidate modalities are different from the first modality, and wherein each of the candidate continuation-inputs references one or more entities represented by the one or more slots; and

presenting, at the client system, one or more suggested inputs corresponding to one or more of the candidate continuation-inputs.

2. The method ofclaim 1, wherein the first modality comprises one of audio, text, image, video, motion, or orientation.

3. The method ofclaim 2, wherein the first modality comprises motion, and wherein the initial input comprises a gesture.

4. The method ofclaim 2, wherein the first modality comprises orientation, and wherein the initial input comprises a gaze on an object.

5. The method ofclaim 1, further comprising:

determining that the first user needs one or more suggested inputs.

6. The method ofclaim 5, wherein determining that the first user needs one or more suggested inputs is based on a wake-up input from the first user.

7. The method ofclaim 6, wherein the wake-up input comprises one or more of a voice utterance, a character string, an image, a video clip, a gesture, or a gaze.

8. The method ofclaim 5, wherein the initial input comprises a gaze on an object, and wherein determining that the first user needs one or more suggested inputs is further based on the gaze on the object.

9. The method ofclaim 8, wherein generating the one or more candidate continuation-inputs is further based on the object.

10. The method ofclaim 5, wherein determining that the first user needs one or more suggested inputs is further based on contextual information associated with the initial input.

11. The method ofclaim 5, wherein determining that the first user needs one or more suggested inputs is further based on the one or more intents.

12. The method ofclaim 1, further comprising:

identifying one or more entities associated with the one or more intents.

13. The method ofclaim 12, wherein generating the one or more candidate continuation-inputs is further based the one or more entities.

14. The method ofclaim 1, further comprising:

receiving, at the client system, a user-selected input from the first user, wherein the user-selected input comprises one of the suggested inputs; and

executing one or more tasks based on the user-selected input.

15. The method ofclaim 1, further comprising:

receiving, at the client system, a first user-selected input from the first user, wherein the first user-selected input comprises one of the suggested inputs, and wherein the first user-selected input is associated with a first intent;

generating, based on the first user-selected input, one or more additional candidate continuation-inputs, wherein each of the one or more additional candidate continuation-inputs is associated with the first intent;

presenting, at the client system, one or more additional suggested inputs corresponding to one or more of the additional candidate continuation-inputs;

receiving, at the client system, a second user-selected input from the first user, wherein the second user-selected input comprises one of the additional suggested inputs; and

executing one or more tasks based on the second user-selected input.

16. The method ofclaim 1, further comprising:

determining the initial input comprises an incomplete input for triggering an execution of one or more tasks corresponding to the one or more intents.

17. The method ofclaim 16, wherein each of the candidate continuation-inputs comprises a complete instruction to trigger the execution of a respective task of the one or more tasks.

18. The method ofclaim 17, wherein each of the presented suggested inputs comprises a guidance to the first user for executing the complete instruction corresponding to the respective candidate continuation-input.

19. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:

receive, at a client system, an initial input from a first user, wherein the initial input is in a first modality;

determine one or more intents and one or more slots corresponding to the initial input, wherein the one or more slots are conditioned on the one or more intents;

generate, based on the one or more intents and the one or more slots, one or more candidate continuation-inputs, where the one or more candidate continuation-inputs are in one or more candidate modalities, respectively, wherein the candidate modalities are different from the first modality, and wherein each of the candidate continuation-inputs references one or more entities represented by the one or more slots; and

present, at the client system, one or more suggested inputs corresponding to one or more of the candidate continuation-inputs.

20. A system comprising: one or more processors; and a non-transitory memory coupled to the processors comprising instructions executable by the processors, the processors operable when executing the instructions to:

receive, at a client system, an initial input from a first user, wherein the initial input is in a first modality;

determine one or more intents and one or more slots corresponding to the initial input, wherein the one or more slots are conditioned on the one or more intents;

generate, based on the one or more intents and the one or more slots, one or more candidate continuation-inputs, where the one or more candidate continuation-inputs are in one or more candidate modalities, respectively, wherein the candidate modalities are different from the first modality, and wherein each of the candidate continuation-inputs references one or more entities represented by the one or more slots; and

present, at the client system, one or more suggested inputs corresponding to one or more of the candidate continuation-inputs.

Images