Movatterモバイル変換


[0]ホーム

URL:


CN112749323B - Method and device for constructing user portrait - Google Patents

Method and device for constructing user portrait
Download PDF

Info

Publication number
CN112749323B
CN112749323BCN201911053500.0ACN201911053500ACN112749323BCN 112749323 BCN112749323 BCN 112749323BCN 201911053500 ACN201911053500 ACN 201911053500ACN 112749323 BCN112749323 BCN 112749323B
Authority
CN
China
Prior art keywords
user
information
commodity
node
nodes
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201911053500.0A
Other languages
Chinese (zh)
Other versions
CN112749323A (en
Inventor
谷育龙
丁卓冶
殷大伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Jingdong Century Trading Co Ltd
Beijing Wodong Tianjun Information Technology Co Ltd
Original Assignee
Beijing Jingdong Century Trading Co Ltd
Beijing Wodong Tianjun Information Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Jingdong Century Trading Co Ltd, Beijing Wodong Tianjun Information Technology Co LtdfiledCriticalBeijing Jingdong Century Trading Co Ltd
Priority to CN201911053500.0ApriorityCriticalpatent/CN112749323B/en
Publication of CN112749323ApublicationCriticalpatent/CN112749323A/en
Application grantedgrantedCritical
Publication of CN112749323BpublicationCriticalpatent/CN112749323B/en
Activelegal-statusCriticalCurrent
Anticipated expirationlegal-statusCritical

Links

Classifications

Landscapes

Abstract

The invention discloses a method and a device for constructing a user portrait, and relates to the technical field of computers. One embodiment of the method comprises the following steps: acquiring a user information heterogram; training to obtain a user portrait construction model for constructing a user portrait based on a graphic neural network according to the user nodes, the commodity information nodes, the first edge and the second edge of the user information heterogram; and adding commodity information related to the user to be constructed into the user information iso-graph, and constructing a user image of the user to be constructed by using the user image construction model. According to the embodiment, the user portrait prediction model can be directly constructed based on various data in the user information heterogram without a manual design mixing method, and a large amount of tag data is not needed, so that the efficiency is high and the cost is low.

Description

Method and device for constructing user portrait
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a method and apparatus for constructing a user portrait.
Background
User portrayal, namely user information tagging (such as gender, age, etc.), is that after collecting and analyzing data of main information such as user static attribute, social attribute, behavior attribute, etc., a basic mode that user's overall view is used for supporting big data applications such as personalized recommendation is abstracted. The user portraits have wider application prospect, especially in the field of electronic commerce, the user population and the equivalent value information of user demands can be rapidly and accurately positioned based on the user portraits, so that the user portraits have important roles in commodity searching, commodity recommending, advertising, accurate marketing and the like.
At present, a common method for constructing user portraits is to construct a user portraits classification model based on artificial design features such as user history behaviors, wherein the user portraits are constructed as a supervised classification task, each user is regarded as an independent data instance, and known user portraits are regarded as labels for supervised learning.
In the process of implementing the present invention, the inventor finds that at least the following problems exist in the prior art: the existing method for constructing the user portrait needs a large number of user marks for supervised learning, and is time-consuming and high in cost; in the process of performing supervised learning, only data such as historical behaviors of the user (such as purchasing goods and clicking goods) are adopted, other data which can be used for constructing the user portrait (such as similarity between the user and other users or similarity of purchasing goods) are not considered, and when multiple data are required to be considered at the same time, a manual design mixing method is required to be used for modeling the multiple data.
Disclosure of Invention
In view of the above, the present invention provides a method and apparatus for constructing a user portrait, which can construct a user portrait prediction model directly based on various data in a user information heterogram without performing a manual design mixing method, and does not require a large amount of tag data, and has high efficiency and low cost.
To achieve the above object, according to one aspect of the present invention, there is provided a method of constructing a user portrait, comprising: acquiring a user information different composition, wherein the nodes of the user information different composition comprise user nodes and commodity information nodes related to the users, a first side of the user information different composition indicates the association between different users, and a second side of the user information different composition indicates the association between the users and the commodity information; training to obtain a user portrait construction model for constructing a user portrait based on a graphic neural network according to the user node, the commodity information node, the first edge and the second edge; and adding commodity information related to the user to be constructed into the user information iso-graph, and constructing a user image of the user to be constructed by using the user image construction model.
Optionally, the god network includes: the system comprises a user information input layer, a commodity information representation layer, a first user representation layer, a second user representation layer and a user portrait construction layer; the user information input layer is used for acquiring the user information iso-graph to acquire the user node, the commodity information node, the first side and the second side; the user information representation layer is used for generating commodity information vectors representing the characteristics of the commodity information nodes; the first user representation layer is used for generating a first user vector representing the characteristics of the user node according to the commodity information vector and the second side; the second user representation layer is used for generating a second user vector representing the characteristics of the user node according to the first user vector and the first edge; the user portrait construction layer is used for constructing the user portrait of the user node according to the second user vector.
Optionally, the node of the user information heterogram further includes: a commodity attribute information node for describing the commodity information; the third side of the user information iso-graph indicates the association between the commodity attribute information and the commodity information.
Optionally, the graph neural network further includes: a commodity attribute information presentation layer; the commodity attribute information representation layer is used for generating commodity attribute information vectors representing the characteristics of the commodity attribute information nodes; and the user information representation layer is used for generating a commodity information vector representing the characteristics of the commodity information node according to the commodity attribute information vector and the third side.
Optionally, a first user node in the user nodes in the user information heterograms has a user portrait tag, and a second user node in the user nodes in the user information heterograms does not have a user portrait tag, and according to the first user node, the second user node, the commodity information node, the first edge and the second edge, a user portrait building model for building the user portrait is obtained through training based on a graphic neural network.
To achieve the above object, according to another aspect of the present invention, there is provided an apparatus for constructing a user portrait, comprising: the system comprises an abnormal composition acquisition module, a construction model acquisition module and a user portrait construction module; the heterogeneous graph acquisition module is used for acquiring a user information heterogeneous graph, wherein the nodes of the user information heterogeneous graph comprise user nodes and commodity information nodes related to the user, a first side of the user information heterogeneous graph indicates the association between different users, and a second side of the user information heterogeneous graph indicates the association between the user and the commodity information; the building model module obtaining module is used for obtaining a user portrait building model for building a user portrait based on training of a graphic neural network according to the user node, the commodity information node, the first edge and the second edge; and the user portrait construction module is used for adding commodity information related to a user to be constructed to the user information iso-composition, and constructing a user portrait of the user to be constructed by using the user portrait construction model.
Optionally, the graph neural network includes: the system comprises a user information input layer, a commodity information representation layer, a first user representation layer, a second user representation layer and a user portrait construction layer; the user information input layer is used for acquiring the user information iso-graph to acquire the user node, the commodity information node, the first side and the second side; the user information representation layer is used for generating commodity information vectors representing the characteristics of the commodity information nodes; the first user representation layer is used for generating a first user vector representing the characteristics of the user node according to the commodity information vector and the second side; the second user representation layer is used for generating a characteristic second user vector representing the user node according to the first user vector and the first edge; the user portrait construction layer is used for constructing the user portrait of the user node according to the second user vector.
Optionally, the node of the user information heterogram further includes: a commodity attribute information node for describing the commodity information; the third side of the user information iso-graph indicates the association between the commodity attribute information and the commodity information.
Optionally, the graph neural network further includes: a commodity attribute information presentation layer; the commodity attribute information representation layer is used for generating characteristic commodity attribute information vectors representing the commodity attribute information nodes; and the user information representation layer is used for generating a commodity information vector representing the characteristics of the commodity information node according to the commodity attribute information vector and the third side.
Optionally, a first user node in the user nodes in the user information heterogram has a user portrait tag, and a second user node in the user nodes in the user information heterogram does not have a user portrait tag; the building model obtaining module is further used for training to obtain a user portrait building model for building the user portrait based on the graphic neural network according to the first user node, the second user node, the commodity information node, the first side and the second side.
To achieve the above object, according to still another aspect of the present invention, there is provided a server for predicting a representation of a user, comprising: one or more processors; a storage means for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement
To achieve the above object, according to still another aspect of the present invention, there is provided a computer-readable medium having stored thereon a computer program, characterized in that the program, when executed by a processor, implements any one of the methods of constructing a user portrait as described above.
The invention has the following advantages or beneficial effects: the nodes and edges of the user information different composition with various nodes are adopted to represent various data such as the user, commodity information, the similarity among different users, the corresponding relation between the user and commodity information and the like, and the nodes and edges of the user information different composition are learned through a graph neural network so as to obtain a user portrait construction model; the method not only realizes simultaneous modeling of various data, but also avoids the problem of manually designing various data mixing methods or manual design characteristics, does not need a large number of data labels, and realizes quick, efficient and low-cost user image construction.
Further effects of the above-described non-conventional alternatives are described below in connection with the embodiments.
Drawings
The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:
FIG. 1 is a schematic diagram of the main flow of a method of constructing a user representation according to an embodiment of the present invention;
FIG. 2a is a schematic diagram of a user information profile according to an embodiment of the present invention;
FIG. 2b is a schematic diagram of another user information profile according to an embodiment of the present invention;
FIG. 2c is a schematic diagram of the neural network of FIG. 2c, according to an embodiment of the present invention;
FIG. 2d is a schematic diagram of another neural network according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of yet another user information profile according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of the major modules of an apparatus for constructing a user representation in accordance with an embodiment of the present invention;
FIG. 5 is an exemplary system architecture diagram in which embodiments of the present invention may be applied;
Fig. 6 is a schematic diagram of a computer system suitable for use in implementing an embodiment of the invention.
Detailed Description
Exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding, and are to be considered merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
As shown in FIG. 1, the embodiment of the invention provides a method for constructing a user portrait, which comprises the following steps:
step S101, a user information different composition is obtained, wherein the nodes of the user information different composition comprise user nodes and commodity information nodes related to the users, the first side of the user information different composition indicates the association between different users, and the second side of the user information different composition indicates the association between the users and the commodity information.
The graph consists of nodes and edges connected among the nodes, and can be divided into isomorphic graphs and heterogeneous graphs according to the number of node types; wherein, isomorphic graph refers to a graph with only one type of node, and heterogeneous graph refers to a graph with two or more types of nodes. The user information heterogram in the embodiment of the application is a graph at least comprising two nodes of user and commodity information, and the commodity information is information of distinguishable commodities such as commodity ID, commodity identification, commodity name and the like. Accordingly, the association between different users indicated by the first edge refers to a similarity relationship between different users, and in the field of electronic commerce, the similarity relationship between users may be calculated based on the similarity of historical behaviors (such as clicking, collecting, purchasing goods, etc.) of two users, for example, if a certain number of goods purchased by both users exceeds a certain number, a first edge may be established between the users, which indicates that the preferences of purchasing goods by both users are relatively similar. The association of the user indicated by the second side with the commodity information means that the historical behavior of the user relates to the commodity, for example, if the user clicks or purchases some commodity, the second side is established between the user and the corresponding commodity information to indicate the tendency of the user to purchase the commodity, i.e. the user is the user who purchases or consumes the commodity.
As shown in fig. 2a, in a preferred embodiment, there is provided a user information profile having two types of nodes, user u and merchandise information i; the user nodes comprise a plurality of user nodes, including u1, u2, u3, u4, u5 and the like, and the user nodes are used for identifying or representing a plurality of commodity information which is purchased, clicked or collected by the user, i1, i2, i3, i4 and the like; the first sides u1u2, u2u5, u3u6, u4u5 and the like represent that the number of commodities purchased or having a purchasing tendency among different users reaches a set threshold (such as 1,2 and the like); the second sides u1i1, u2i1, u3i2, etc. represent products that the user purchases or has a tendency to purchase. Specifically, taking the user node u2 as an example, the user u2 purchases or tends to purchase the product i1, and thus the user u2 and the product i1 have the second side u2i1 therebetween, and the user u2 and the user u1 have the similar relationship, and thus the first side u1u2 therebetween, because the user u2 and the user u1 both purchase or tend to purchase the product i 1.
It can be understood that, since commodity information (such as a mobile phone) has commodity attribute information, such as color, brand, model and the like, which can be used for describing the commodity information, in the process of constructing the user information iso-composition, the user information iso-composition can be constructed according to various information, such as the user, commodity information, commodity attribute information and the like, so that the establishment of a user portrait construction model can be realized based on more various information, and a user portrait can be constructed more comprehensively, stereoscopically and accurately.
In an alternative embodiment, the node of the user information profile further includes: a commodity attribute information node for describing the commodity information; the third side of the user information iso-graph indicates the association between the commodity attribute information and the commodity information.
In particular, referring to fig. 2b, the node of the user information profile constructed in a preferred embodiment comprises: user node u, commodity information node i and commodity attribute information node t; the user nodes are provided with a plurality of user nodes, including u1, u2, u3, u4, u5 and the like, and are used for identifying or representing commodity information such as i1, i2, i3, i4 and the like which are purchased, clicked or collected by the user, and commodity attribute information used for describing the commodity information comprises t1, t2, t3, t4 and the like; the first sides u1u2, u2u5, u3u6, u4u5 and the like represent that the number of commodities purchased or having a purchasing tendency among different users reaches a set threshold (such as 1,2 and the like); the second sides u1i1, u2i1, u3i2, etc. represent products purchased or having a tendency to be purchased by the user; the third sides i1t1, i1t2, i2t2, and the like represent product attribute information included in the product information. Specifically, taking the user node u2 as an example, since the user u2 purchases or tends to purchase the commodity i1, a second side u2i1 is provided between the user u2 and the commodity i1; meanwhile, since both the user u2 and the user u1 purchase or tend to purchase the commodity i1, the user u2 and the user u1 have a similar relationship, and thus have a first edge u1u2; since the commodity information i1 has both the commodity attribute information of t1 and t2, the commodity information i1 has both the third sides i1t1 and i1t2.
Step S102, training to obtain a user portrait construction model for constructing a user portrait based on a graphic neural network according to the user node, the commodity information node, the first side and the second side.
In an alternative embodiment, the god network comprises: the system comprises a user information input layer, a commodity information representation layer, a first user representation layer, a second user representation layer and a user portrait construction layer; the user information input layer is used for acquiring the user information iso-graph to acquire the user node, the commodity information node, the first side and the second side; the user information representation layer is used for generating commodity information vectors representing the characteristics of the commodity information nodes; the first user representation layer is used for generating a first user vector representing the characteristics of the user node according to the commodity information vector and the second side; the second user representation layer is used for generating a second user vector representing the characteristics of the user node according to the first user vector and the first edge; the user portrait construction layer is used for constructing the user portrait of the user node according to the second user vector.
Specifically, referring to fig. 2a and 2c, after receiving the user information iso-graph shown in fig. 2a, the user information input layer of the neural network (shown in fig. 2 c) acquires user nodes indicated by the user information iso-graph, such as u1, u2, u3, u4, u5, commodity information nodes indicated by the user information iso-graph, such as i1, i2, i3, i4, commodity attribute information nodes indicated by t1, t2, t3, t4, and the like, and first edges u1u2, u2u5, u3u6, u4u5, and the like, second edges u1i1, u2i1, u3i2, and the like, and a third edge. The commodity information representing layer is configured to represent each commodity information as a low-dimensional vector composed of a plurality of real numbers, for example, the commodity information vector corresponding to the commodity information i1 may be a low-dimensional vector with a length of T: s= [ s1,s2,…,sT ], where si is a real number. After the first user representation layer receives the commodity information vector, the second edges u1i1, u2i1, u3i2 and the like transmitted by the commodity information representation layer, each user may purchase or tend to purchase various commodities, so that the weight corresponding to the commodity information vector representing the first user is determined through an attention mechanism according to the second edges of all the user nodes and the corresponding commodity information vector, and the weight of the commodity information vector is used for weighting and summing to obtain the corresponding first user vector, such as u= [ v1,v2,…,vT ], wherein vi is a real number. After receiving the first user vector, the first edges u1u2, u2u5, u3u6, u4u5, and so on, the second user representation layer determines the weight of the first user vector through the attention mechanism according to all the second edges of each user and the corresponding first user representation, and obtains the corresponding second user vector through the weighted summation of the weights of the first user vector, such as u= [ u1,u2,…,uT ], wherein ui is a real number. On the basis, a user portrait construction layer constructs a user portrait corresponding to each user based on the second user vector of the user.
In an alternative embodiment, the node of the user information profile further includes: a commodity attribute information node for describing the commodity information; in the case where the third side of the user information iso-graph indicates the association between the commodity attribute information and the commodity information, the graph neural network further includes: a commodity attribute information presentation layer; the commodity attribute information representation layer is used for generating commodity attribute information vectors representing the characteristics of the commodity attribute information nodes; and the user information representation layer is used for generating a commodity information vector representing the characteristics of the commodity information node according to the commodity attribute information vector and the third side.
Referring to fig. 2b and 2d, after receiving the user information iso-graph shown in fig. 2b, the user information input layer of the neural network (as shown in fig. 2 d) acquires user nodes such as u1, u2, u3, u4 indicated by the user information iso-graph, commodity information nodes such as i1, i2, i3, i4, and the like, first sides u1u2, u2u5, u3u6, u4u5, second sides u1i1, u2i1, u3i2, and the like, and third sides i1t1, i1t2, i2t2, and the like. The commodity attribute information representation layer is configured to represent each commodity attribute information as a low-dimensional vector composed of a plurality of real numbers, for example, the commodity attribute information vector corresponding to the commodity attribute information T1 may be a low-dimensional vector with a length of T: e= [ e1,e2,…,eT ], where ei is a real number. After receiving the commodity attribute information vector, the third sides i1t1, i1t2, i2t2, etc., the commodity information expression layer determines the weight of each commodity attribute information vector through the attention mechanics learning mechanism according to all the third sides of each commodity information and the corresponding commodity attribute information, and determines the commodity information vector corresponding to the commodity information through the weight weighted summation of the commodity attribute information vectors, such as s= [ s1,s2,…,sT ], Where si is a real number. After the first user representation layer receives the commodity information vector, the second edges u1i1, u2i1, u3i2 and the like transmitted by the commodity information representation layer, as each user may purchase or tends to purchase multiple commodities, the weight corresponding to the commodity information vector representing the first user is determined through the attention mechanism according to the second edges of all the user nodes and the corresponding commodity information vector, and the corresponding first user vector is obtained through the weighted summation of the weights of the commodity information vectors, such as u= [ v1,v2,…,vT ], Where vi is a real number. After receiving the first user vector, the first edges u1u2, u2u5, u3u6, u4u5, and so on, the second user representation layer determines the weight of the first user vector through the attention mechanism according to all the second edges of each user and the corresponding first user representation, and obtains the corresponding second user vector through the weighted summation of the weights of the first user vector, such as u= [ u1,u2,…,uT ], wherein ui is a real number. on the basis, a user portrait construction layer constructs a user portrait corresponding to each user based on the second user vector of the user.
Furthermore, a first user node in the user nodes in the user information heterograms is provided with a user portrait tag, and a second user node in the user nodes in the user information heterograms is not provided with a user portrait tag, and a user portrait building model for building the user portrait is obtained through training based on a graphic neural network according to the first user node, the second user node, the commodity information node, the first edge and the second edge. Specifically, parameters of a user information input layer, a commodity information representation layer, a first user representation layer, a second user representation layer, a user portrait construction layer and the like in the graphic neural network are trained and optimized by comparing the constructed user portraits with consistent user portrait labels and calculating a loss function through cross entropy. Thus, a model can be built based on the trained user portraits to build a user portraits for each user in the user information profile.
Referring to fig. 3, in a preferred embodiment, a user information profile containing new users is provided. Specifically, the user figures of the users u1, u2, u5, and u7 are exemplified as male, female, and female, respectively. In the user portrayal construction process, we consider the portrayal of users u2 and u7 as unknown. For both u1 and u2 users, the commodities i1 and i3 are purchased, the purchasing habits are similar, and the user portrait of the user u2 constructed based on the model is male. For both u5 and u7 users, only one and the same item i5 was purchased. Since the two items i5 and i6 are purchased by female user u7, they may be more female-oriented items, and thus the corresponding item attributes t6 and t7, may also be more female-oriented. Since the commodity i4 includes the attribute t6, it is determined that the commodity i4 is more female, and based on the characteristics of both the commodities i5 and i4 being more female, a user portrait of the user u5 is constructed as female. Meanwhile, if the user portraits corresponding to u2 and u7 of the model construction are inconsistent with the known user portraits, the user portraits construction model is continuously optimized based on the arithmetic function.
And step S103, adding commodity information related to the user to be constructed to the user information iso-composition, and constructing a user image of the user to be constructed by using the user image construction model.
It can be understood that the user portrait construction model obtained after the user information heterogram learning can realize the construction of the user portrait corresponding to the user in the user information heterogram. Therefore, when a user representation of a new user needs to be built, the new user, corresponding commodity information, commodity attribute information and the like need to be added to the user information heterogram as nodes, corresponding first edges, second edges, third edges and the like are built for building, and then the user representation of the new user is built based on the user information heterogram containing the new user.
The invention has the following advantages or beneficial effects: the nodes and edges of the user information different composition with various nodes are adopted to represent various data such as the user, commodity information, the similarity among different users, the corresponding relation between the user and commodity information and the like, and the nodes and edges of the user information different composition are learned through a graph neural network so as to obtain a user portrait construction model; the method not only realizes simultaneous modeling of various data, but also avoids the problem of manually designing various data mixing methods or manual design characteristics, does not need a large number of data labels, and realizes quick, efficient and low-cost user image construction.
Referring to fig. 4, in accordance with the above embodiment, there is provided an apparatus 400 for constructing a user portrait, which includes: the system comprises an abnormal composition acquisition module 401, a construction model acquisition module 402 and a user portrait construction module 403; wherein,
The heterogeneous map obtaining module 401 is configured to obtain a user information heterogeneous map, where a node of the user information heterogeneous map includes a user node and a commodity information node related to the user, a first edge of the user information heterogeneous map indicates a relationship between different users, and a second edge of the user information heterogeneous map indicates a relationship between the user and the commodity information; the building model module obtaining module 402 is configured to train to obtain a user portrait building model for building a user portrait based on a graph neural network according to the user node, the commodity information node, the first edge, and the second edge; the user portrait construction module 403 is configured to add merchandise information related to a user to be constructed to the user information iso-composition, and construct a user portrait of the user to be constructed by using the user portrait construction model.
In an alternative embodiment, the graph neural network includes: the system comprises a user information input layer, a commodity information representation layer, a first user representation layer, a second user representation layer and a user portrait construction layer; the user information input layer is used for acquiring the user information iso-graph to acquire the user node, the commodity information node, the first side and the second side; the user information representation layer is used for generating commodity information vectors representing the characteristics of the commodity information nodes; the first user representation layer is used for generating a first user vector representing the characteristics of the user node according to the commodity information vector and the second side; the second user representation layer is used for generating a second user vector representing the characteristics of the user node according to the first user vector and the first edge; the user portrait construction layer is used for constructing the user portrait of the user node according to the second user vector.
In an alternative embodiment, the node of the user information profile further includes: a commodity attribute information node for describing the commodity information; the third side of the user information iso-graph indicates the association between the commodity attribute information and the commodity information.
In an alternative embodiment, the graph neural network further includes: a commodity attribute information presentation layer; the commodity attribute information representation layer is used for generating commodity attribute information vectors representing the characteristics of the commodity attribute information nodes; and the user information representation layer is used for generating a commodity information vector representing the characteristics of the commodity information node according to the commodity attribute information vector and the third side.
In an alternative embodiment, a first user node of the user nodes in the user information profile has a user portrayal tag and a second user node of the user nodes in the user information profile does not have a user portrayal tag; the build model obtaining module 402 is further configured to train to obtain a user portrait build model for building a user portrait based on a graphic neural network according to the first user node, the second user node, the commodity information node, the first edge, and the second edge.
FIG. 5 illustrates an exemplary system architecture 500 in which the user portrayal construction method or user portrayal construction device of embodiments of the present invention may be applied.
As shown in fig. 5, the system architecture 500 may include terminal devices 501, 502, 503, a network 504, and a server 505. The network 504 is used as a medium to provide communication links between the terminal devices 501, 502, 503 and the server 505. The network 504 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.
A user may interact with the server 505 via the network 504 using the terminal devices 501, 502, 503 to receive or send messages or the like. Various communication client applications, such as shopping class applications, web browser applications, search class applications, instant messaging tools, mailbox clients, social platform software, etc., may be installed on the terminal devices 501, 502, 503.
The terminal devices 501, 502, 503 may be a variety of electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablets, laptop and desktop computers, and the like.
The server 505 may be a server providing various services, such as a background management server providing support for shopping-type websites browsed by the user using the terminal devices 501, 502, 503. The background management server can analyze and other data such as the received product information inquiry request and feed back the processing result (constructed user portrait) to the terminal equipment.
It should be noted that, the method for building a user portrait according to the embodiment of the present invention is generally executed by the server 505, and accordingly, the device for building a user portrait is generally disposed in the server 505.
It should be understood that the number of terminal devices, networks and servers in fig. 5 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
Referring now to FIG. 6, there is illustrated a schematic diagram of a computer system 600 suitable for use in implementing an embodiment of the present invention. The terminal device shown in fig. 6 is only an example, and should not impose any limitation on the functions and the scope of use of the embodiment of the present invention.
As shown in fig. 6, the computer system 600 includes a Central Processing Unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 602 or a program loaded from a storage section 608 into a Random Access Memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, ROM 602, and RAM 603 are connected to each other through a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, mouse, etc.; an output portion 607 including a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, a speaker, and the like; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. The drive 610 is also connected to the I/O interface 605 as needed. Removable media 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is installed as needed on drive 610 so that a computer program read therefrom is installed as needed into storage section 608.
In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication portion 609, and/or installed from the removable medium 611. The above-described functions defined in the system of the present invention are performed when the computer program is executed by a Central Processing Unit (CPU) 601.
The computer readable medium shown in the present invention may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The modules involved in the embodiments of the present invention may be implemented in software or in hardware. The described modules may also be provided in a processor, for example, as: a processor comprises a heterogeneous graph acquisition module, a construction model acquisition module and a user portrait construction module. The names of these modules do not limit the module itself in some cases, and for example, the heterogeneous map acquisition module may also be described as a "module for acquiring a user information heterogeneous map".
As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be present alone without being fitted into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to include obtaining a user information profile, the user information profile including nodes including user nodes, merchandise information nodes associated with the user, a first side of the user information profile indicating an association between different ones of the users, a second side of the user information profile indicating an association between the user and the merchandise information; training to obtain a user portrait construction model for constructing a user portrait based on a graphic neural network according to the user node, the commodity information node, the first edge and the second edge; and adding commodity information related to the user to be constructed into the user information iso-graph, and constructing a user image of the user to be constructed by using the user image construction model.
According to the technical scheme of the embodiment of the invention, as the nodes and edges of the user information heterograms with various nodes are adopted to represent various data such as users, commodity information, similarity among different users, corresponding relation between the users and commodity information and the like, the nodes and edges of the user information heterograms are learned through the graph neural network so as to obtain a user portrait construction model; the method not only realizes simultaneous modeling of various data, but also avoids the problem of manually designing various data mixing methods or manual design characteristics, does not need a large number of data labels, and realizes quick, efficient and low-cost user image construction.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives can occur depending upon design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.

Claims (12)

CN201911053500.0A2019-10-312019-10-31Method and device for constructing user portraitActiveCN112749323B (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN201911053500.0ACN112749323B (en)2019-10-312019-10-31Method and device for constructing user portrait

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN201911053500.0ACN112749323B (en)2019-10-312019-10-31Method and device for constructing user portrait

Publications (2)

Publication NumberPublication Date
CN112749323A CN112749323A (en)2021-05-04
CN112749323Btrue CN112749323B (en)2024-07-23

Family

ID=75645069

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN201911053500.0AActiveCN112749323B (en)2019-10-312019-10-31Method and device for constructing user portrait

Country Status (1)

CountryLink
CN (1)CN112749323B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN112465565B (en)*2020-12-112023-09-26加和(北京)信息科技有限公司User portrait prediction method and device based on machine learning
CN113378051B (en)*2021-06-162024-03-22南京大学 Crowdsourcing task recommendation method based on user-task association based on graph neural network
CN114238760B (en)*2021-12-152025-05-23浙江网商银行股份有限公司 Project recommendation method and device
CN115883147B (en)*2022-11-222023-10-13浙江御安信息技术有限公司Attacker portrait method based on graphic neural network

Citations (2)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN106874435A (en)*2017-01-252017-06-20北京航空航天大学User portrait construction method and device
CN107464141A (en)*2017-08-072017-12-12北京京东尚科信息技术有限公司For the method, apparatus of information popularization, electronic equipment and computer-readable medium

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
KR101443158B1 (en)*2013-07-112014-09-25오드컨셉 주식회사Commodity information recommending system based on user interest
CN105824912A (en)*2016-03-152016-08-03平安科技(深圳)有限公司Personalized recommending method and device based on user portrait
CN109993966A (en)*2018-01-022019-07-09中国移动通信有限公司研究院 A method and device for constructing a user portrait
CN108229590B (en)*2018-02-132020-05-15阿里巴巴集团控股有限公司Method and device for acquiring multi-label user portrait
CN110309405B (en)*2018-03-082022-09-30腾讯科技(深圳)有限公司Project recommendation method and device and storage medium
CN109242633B (en)*2018-09-202022-04-08创新先进技术有限公司Commodity pushing method and device based on bipartite graph network
CN109903117B (en)*2019-01-042022-10-25苏宁易购集团股份有限公司Knowledge graph processing method and device for commodity recommendation

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN106874435A (en)*2017-01-252017-06-20北京航空航天大学User portrait construction method and device
CN107464141A (en)*2017-08-072017-12-12北京京东尚科信息技术有限公司For the method, apparatus of information popularization, electronic equipment and computer-readable medium

Also Published As

Publication numberPublication date
CN112749323A (en)2021-05-04

Similar Documents

PublicationPublication DateTitle
CN112749323B (en)Method and device for constructing user portrait
CN113763093B (en)Article recommending method and device based on user portrait
CN113450172B (en)Commodity recommendation method and device
CN110619078B (en)Method and device for pushing information
CN112925973B (en)Data processing method and device
CN105208113A (en)Information pushing method and device
CN110020162B (en)User identification method and device
CN112990954B (en)Coupon distribution method and device
CN112950321B (en) Item recommendation method and device
CN107274209A (en)The method and apparatus for predicting advertising campaign sales data
CN113516524B (en) Method and device for pushing information
CN110866040B (en)User portrait generation method, device and system
CN113742564B (en) Method and device for pushing target resources
CN113450167A (en)Commodity recommendation method and device
CN113778979B (en) A method and device for determining click rate of live broadcast
CN112449217B (en)Method and device for pushing video, electronic equipment and computer readable medium
CN113822734B (en)Method and device for generating information
US20120265588A1 (en)System and method for recommending new connections in an advertising exchange
CN113159877B (en)Data processing method, device, system and computer readable storage medium
CN113269600B (en)Information sending method and device
CN113327145B (en)Article recommendation method and device
CN111833085B (en)Method and device for calculating price of article
CN111784091B (en)Method and device for processing information
CN112819619A (en)Transaction processing method and device
CN113327147A (en)Method and device for displaying article information

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
GR01Patent grant
GR01Patent grant

[8]ページ先頭

©2009-2025 Movatter.jp