Disclosure of Invention
The application provides an advertisement recommending method, an advertisement recommending system, electronic equipment and an advertisement recommending device, which are beneficial to improving the recommending effect of an advertisement platform on search advertisements, so that the recommended advertisements are more in line with the intention of users, and the satisfaction degree of the users can be improved.
The application provides an advertisement recommendation method, which is applied to first equipment and comprises the steps of receiving an advertisement request sent by second equipment, wherein the advertisement request comprises search content, determining a target value type corresponding to the advertisement request based on the search content in the advertisement request, recommending the advertisement request based on the target value type corresponding to the advertisement request to obtain N search advertisements corresponding to the target value type, wherein N is a positive integer, and sending the N search advertisements to the second equipment.
In the application, the first equipment identifies the value type of the advertisement request after receiving the advertisement request, so that the advertisement request corresponding to the identified value type can be pertinently recommended, the recommended advertisement is more in line with the intention of the user, and the satisfaction degree of the user is improved.
In one possible implementation manner, the method further comprises the step of taking the result of determining the target value type corresponding to the advertisement request as a virtual advertisement space.
According to the application, the first equipment sets different virtual advertisement positions for advertisements recommended by the advertisement requests with different value types, and can be displayed to an advertiser for viewing, so that the advertiser can intuitively and specifically know the value of the advertisement requests. In addition, the advertisement platform can set different bidding strategies for different virtual advertisement positions, so that advertisers can bid based on different virtual advertisement positions, and the income of the advertisement platform is improved.
In one possible implementation manner, the advertisement recommendation is performed on the advertisement request based on the target value type corresponding to the advertisement request, and the N search advertisements corresponding to the target value type are obtained, wherein the N search advertisements corresponding to the target value type are obtained by determining the target advertisement type corresponding to the advertisement request based on the search content in the advertisement request, and the advertisement recommendation is performed on the advertisement request based on the target value type corresponding to the advertisement request and the target advertisement type, and the N search advertisements corresponding to the target value type and the target advertisement type are obtained.
In the application, the first equipment can recommend the advertisement corresponding to the advertisement type through identifying the advertisement type required by the user, so that the recommended advertisement is more in line with the intention of the user, and the satisfaction degree of the user is improved.
In one possible implementation manner, the determining the target advertisement type corresponding to the advertisement request based on the search content in the advertisement request comprises determining a user intention based on the search content in the advertisement request, and determining the target advertisement type corresponding to the advertisement request based on the user intention.
In one possible implementation, the advertisement types include at least application advertisements and merchandise advertisements.
In one possible implementation, the N search advertisements are top-ranked N search advertisements, and the ranking of the N search advertisements is determined by estimating thousands of presentation benefits ECPM.
In one possible implementation manner, the N search advertisements are obtained after ranking based on a first search advertisement and a second search advertisement, where the first search advertisement is obtained after advertisement recommendation in the first device, and the second search advertisement is obtained after advertisement recommendation in a third device, and the third device is a device of a third party requester platform DSP.
In the application, the first equipment comprehensively considers the ranking of the recommended advertisements of the DSP and the ranking of the recommended advertisements of the third-party DSP, so that the finally output search advertisements with the top N ranking can consider the ranking of the multi-party DSP, and the errors caused by the recommendation of the single-party DSP are reduced.
In one possible implementation manner, different value types correspond to different recommendation models, the advertisement request is recommended based on a target value type corresponding to the advertisement request to obtain N search advertisements corresponding to the target value type, the method comprises the steps of determining a target recommendation model corresponding to the target value type based on the target value type corresponding to the advertisement request, and using the target recommendation model corresponding to the target value type to recommend the advertisement request to obtain N search advertisements corresponding to the target value type.
In the application, the first equipment uses different recommendation models to recommend advertisements of different price types, thereby being capable of rapidly and accurately processing the advertisements of different price types to obtain the required recommended advertisements.
In one possible implementation manner, the determining the target value type corresponding to the advertisement request based on the search content in the advertisement request includes identifying the search content in the advertisement request to obtain a keyword, and text matching the keyword with an advertisement list to determine the target value type corresponding to the advertisement request.
In the application, the first equipment can quickly determine the value type corresponding to the advertisement request by matching the keywords in the search content with the text in the advertisement list.
In one possible implementation, the method further includes periodically updating the in-flight advertising list.
In the application, the first equipment can avoid errors caused by the change of the advertisement casting list by periodically updating the advertisement casting list.
In one possible implementation manner, the text matching includes a plurality of results, the plurality of results and the plurality of value types are in one-to-one mapping relationship, or the plurality of results and the plurality of value types are in many-to-one mapping relationship.
The application provides an advertisement recommendation method, which is applied to an advertisement recommendation system, and comprises first equipment and second equipment, wherein the second equipment responds to search content input by a user and sends an advertisement request to the first equipment, the first equipment receives the advertisement request sent by the second equipment, the advertisement request comprises the search content, the first equipment determines a target value type corresponding to the advertisement request based on the search content in the advertisement request, the first equipment carries out advertisement recommendation on the advertisement request based on the target value type corresponding to the advertisement request to obtain N search advertisements corresponding to the target value type, N is a positive integer, the first equipment sends the N search advertisements to the second equipment, and the second equipment displays the N search advertisements.
In the application, after the second equipment responds to the search content input by the user and sends the advertisement request to the first equipment, the first equipment receives the advertisement request and identifies the value type of the advertisement request, so that the advertisement request corresponding to the identified value type can be pertinently recommended by the second equipment, the recommended advertisement is more in line with the intention of the user, and the recommended advertisement is presented to the user on the second equipment so as to improve the satisfaction degree of the user.
In one possible implementation manner, the method further comprises the step that the first device takes a result of determining the target value type corresponding to the advertisement request as a virtual advertisement space.
According to the application, the first equipment sets different virtual advertisement positions for advertisements recommended by the advertisement requests with different value types, and can be displayed to an advertiser for viewing, so that the advertiser can intuitively and specifically know the value of the advertisement requests. In addition, the advertisement platform can set different bidding strategies for different virtual advertisement positions, so that advertisers can bid based on different virtual advertisement positions, and the income of the advertisement platform is improved.
In one possible implementation manner, the first device performs advertisement recommendation on the advertisement request based on the target value type corresponding to the advertisement request to obtain N search advertisements corresponding to the target value type, and the method comprises the steps that the first device determines a target advertisement type corresponding to the advertisement request based on the search content in the advertisement request, and performs advertisement recommendation on the advertisement request based on the target value type corresponding to the advertisement request and the target advertisement type to obtain the N search advertisements corresponding to the target value type and the target advertisement type.
In the application, the first equipment can recommend the advertisement corresponding to the advertisement type through identifying the advertisement type required by the user, so that the recommended advertisement is more in line with the intention of the user, and the satisfaction degree of the user is improved.
In one possible implementation manner, the first device determines a target advertisement type corresponding to the advertisement request based on the search content in the advertisement request, and the first device determines a user intention based on the search content in the advertisement request, and determines the target advertisement type corresponding to the advertisement request based on the user intention.
In the application, the first equipment can recommend the advertisement corresponding to the advertisement type through identifying the advertisement type required by the user, so that the recommended advertisement is more in line with the intention of the user, and the satisfaction degree of the user is improved.
In one possible implementation, the advertisement types include at least application advertisements and merchandise advertisements.
In one possible implementation, the N search advertisements are top-ranked N search advertisements, and the ranking of the N search advertisements is determined by estimating thousands of presentation benefits ECPM.
In one possible implementation manner, the N search advertisements are obtained after ranking based on a first search advertisement and a second search advertisement, where the first search advertisement is obtained after advertisement recommendation in the first device, and the second search advertisement is obtained after advertisement recommendation in a third device, and the third device is a device of a third party requester platform DSP.
In the application, the first equipment comprehensively considers the ranking of the recommended advertisements of the DSP and the ranking of the recommended advertisements of the third-party DSP, so that the finally output search advertisements with the top N ranking can consider the ranking of the multi-party DSP, and the errors caused by the recommendation of the single-party DSP are reduced.
In one possible implementation manner, different value types correspond to different recommendation models, the advertisement request is recommended based on a target value type corresponding to the advertisement request to obtain N search advertisements corresponding to the target value type, the method comprises the steps of determining a target recommendation model corresponding to the target value type based on the target value type corresponding to the advertisement request, and using the target recommendation model corresponding to the target value type to recommend the advertisement request to obtain N search advertisements corresponding to the target value type.
In the application, the first equipment uses different recommendation models to recommend advertisements of different price types, thereby being capable of rapidly and accurately processing the advertisements of different price types to obtain the required recommended advertisements.
In one possible implementation manner, the determining the target value type corresponding to the advertisement request based on the search content in the advertisement request includes identifying the search content in the advertisement request to obtain a keyword, and text matching the keyword with an advertisement list to determine the target value type corresponding to the advertisement request.
In the application, the first equipment can quickly determine the value type corresponding to the advertisement request by matching the keywords in the search content with the text in the advertisement list.
In one possible implementation, the method further includes periodically updating the in-flight advertising list.
In the application, the first equipment can avoid errors caused by the change of the advertisement casting list by periodically updating the advertisement casting list.
In one possible implementation manner, the text matching includes a plurality of results, the plurality of results and the plurality of value types are in one-to-one mapping relationship, or the plurality of results and the plurality of value types are in many-to-one mapping relationship.
In a third aspect, the present application provides an advertisement recommendation system, including a first device and a second device, where the first device implements the advertisement recommendation method as performed by the first device in the second aspect, or the first device implements the advertisement recommendation method as described in the first aspect, and the second device implements the advertisement recommendation method as performed by the second device in the second aspect.
In a fourth aspect, the application provides an electronic device comprising a memory, one or more processors and one or more programs, wherein the one or more programs are stored in the memory, and the one or more processors, when executing the one or more programs, cause the electronic device to implement the advertisement recommendation method according to the first aspect.
In a fifth aspect, the present application provides an advertisement recommendation apparatus comprising one or more functional modules for performing any of the advertisement recommendation methods provided in the first aspect.
In a sixth aspect, the present application provides a chip system, the chip system comprising a processing circuit, a storage medium having stored therein computer program code, which when executed by the processing circuit implements the advertisement recommendation method according to the first or second aspect.
In a seventh aspect, the present application provides a readable storage medium having stored therein a program which, when run on a first device, causes the first device to implement the advertisement recommendation method according to the first or second aspect.
In an eighth aspect, the present application provides a program which, when run on a processor from a first device, causes the first device to perform the advertisement recommendation method as described in the first aspect.
In one possible design, the program in the eighth aspect may be stored in whole or in part on a storage medium packaged with the processor, or in part or in whole on a memory not packaged with the processor.
Detailed Description
In the embodiment of the present application, unless otherwise specified, the character "/" indicates that the associated object is one or the relationship. For example, A/B may represent A or B. "and/or" describes an association relationship of an association object, meaning that three relationships may exist. For example, A and/or B may mean that A alone, both A and B, and B alone are present.
It should be noted that the terms "first," "second," and the like in the embodiments of the present application are used for distinguishing between description and not necessarily for indicating or implying a relative importance or number of features or characteristics in order.
In the embodiments of the present application, "at least one" means one or more, and "a plurality" means two or more. Furthermore, "at least one item(s)" below, or the like, refers to any combination of these items, and may include any combination of single item(s) or plural items(s). For example, at least one of A, B or C may represent A, B, C, A and B, A and C, B and C, or A, B and C. Wherein each of A, B, C may itself be an element or a collection of one or more elements.
In embodiments of the application, "exemplary," "in some embodiments," "in another embodiment," etc. are used to indicate an example, instance, or illustration. Any embodiment or design described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, the term use of an example is intended to present concepts in a concrete fashion.
"Of", "corresponding (corresponding, relevant)" and "corresponding (corresponding)" in the embodiments of the present application may be sometimes mixed, and it should be noted that the meanings to be expressed are consistent when the distinction is not emphasized. In the embodiments of the present application, communications and transmissions may sometimes be mixed, and it should be noted that, when the distinction is not emphasized, the meaning expressed is consistent. For example, a transmission may include sending and/or receiving, either nouns or verbs.
The equal to that related in the embodiment of the application can be used together with the greater than the adopted technical scheme, can also be used together with the lesser than the adopted technical scheme. It should be noted that when the combination is equal to or greater than the combination, the combination cannot be used with less than the combination, and when the combination is equal to or less than the combination, the combination cannot be used with greater than the combination.
The advertisement search (SEARCH ADVERTISEMENTS, SEARCH ADS) refers to an advertisement link mode that after an advertiser puts advertisement materials and keywords through an advertisement platform, a user can actively input search word inquiry and then return related advertisements.
It can be understood that the search advertisement is an advertisement obtained by searching in the advertisement platform by the user in an advertisement searching mode, and the display advertisement is an advertisement actively displayed by the advertisement platform. When the search terms of the users are different, the intentions of the users for the advertisements are also different. In addition, the search advertisement has a higher correlation with the user's intention than the display advertisement, and thus has a higher value than the display advertisement, which can be reflected in a higher average thousand display revenue (CPM).
However, in the current advertisement platform in the process of recommending search advertisements, because the advertisement recommended by the advertisement platform has low matching degree with the intention of the user and poor recommendation effect, the advertisement platform is difficult to identify high-value advertisements, and a reasonable bidding strategy cannot be set for the high-value advertisements, so that insufficient advertisement bidding is caused, the advertisement platform is lost, and the satisfaction degree of advertisers is reduced.
Based on the above problems, the embodiment of the application provides an advertisement recommendation method which is applied to electronic equipment. The electronic device may be a server in an advertisement platform, where the server may be a windows server, a Linux server, or other server devices that may provide multiple devices for simultaneous access, and the embodiment of the present application does not specifically limit the type of the server.
In some alternative embodiments, the electronic device may also be a server cluster in the advertisement platform, where the server cluster may be a device cluster formed by multiple regions, multiple machine rooms, and multiple servers, and the embodiments of the present application are not limited in particular.
It can be understood that the electronic device in the embodiment of the application can support the functions of message storage and distribution, large-scale data storage, large-scale data processing, multiple concurrent high-concurrency processing, data redundancy backup and the like.
Fig. 1 is a schematic diagram schematically illustrating a structure of an electronic device 100.
The electronic device 100 may include at least one processor and at least one memory communicatively coupled to the processor, wherein the memory stores program instructions executable by the processor, and wherein the processor is capable of executing the methods provided by the embodiments shown herein when the processor invokes the program instructions.
Fig. 1 illustrates a block diagram of an exemplary electronic device 100 suitable for implementing embodiments herein. The electronic device 100 shown in fig. 1 is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments herein.
As shown in fig. 1, components of electronic device 100 may include, but are not limited to, one or more processors 110, a memory 120, a communication bus 140 that connects the various system components (including memory 120 and processor 110), and a communication interface 130.
Communication bus 140 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include industry Standard architecture (Industry Standard Architecture; hereinafter ISA) bus, micro channel architecture (Micro Channel Architecture; hereinafter MAC) bus, enhanced ISA bus, video electronics standards Association (Video Electronics Standards Association; hereinafter VESA) local bus, and peripheral component interconnect (PERIPHERAL COMPONENT INTERCONNECTION; hereinafter PCI) bus.
Electronic device 100 typically includes multiple computer system readable media. Such media can be any available media that can be accessed by the device and includes both volatile and nonvolatile media, removable and non-removable media.
Memory 120 may include computer system readable media in the form of volatile memory, such as random access memory (Random Access Memory; hereinafter: RAM) and/or cache memory. The device may further include other removable/non-removable, volatile/nonvolatile computer system storage media. Although not shown in FIG. 1, a disk drive for reading from and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable non-volatile optical disk (e.g., a compact disk read only memory (Compact Disc Read Only Memory; hereinafter CD-ROM), digital versatile disk read only memory (Digital Video Disc Read Only Memory; hereinafter DVD-ROM), or other optical media) may be provided. In these cases, each drive may be coupled to communication bus 140 through one or more data medium interfaces. Memory 120 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments herein.
A program/utility having a set (at least one) of program modules may be stored in the memory 120, such program modules include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment. Program modules typically carry out the functions and/or methods of the embodiments described herein.
Electronic device 100 may also communicate with one or more external devices (e.g., keyboard, pointing device, display, etc.), one or more devices that enable a user to interact with the device, and/or any device (e.g., network card, modem, etc.) that enables the device to communicate with one or more other devices. Such communication may occur through the communication interface 130. Moreover, electronic device 100 may also communicate with one or more networks (e.g., local area network (Local Area Network; hereinafter: LAN), wide area network (Wide Area Network; hereinafter: WAN) and/or a public network, such as the Internet) via a network adapter (not shown in FIG. 1) that may communicate with other modules of the device via communication bus 140. It should be appreciated that although not shown in FIG. 1, other hardware and/or software modules may be used in connection with electronic device 100, including, but not limited to, microcode, device drivers, redundant processing units, external disk drive arrays, disk array (Redundant Arrays of INDEPENDENT DRIVES; RAID) systems, tape drives, data backup storage systems, and the like.
The processor 110 executes various functional applications and data processing, such as implementing the methods provided by the embodiments herein, by running programs stored in the memory 120.
It should be understood that the interfacing relationship between the modules illustrated in the embodiments herein is only illustrative and not limiting on the structure of the electronic device 100. In other embodiments herein, the electronic device 100 may also employ different interfaces in the above embodiments, or a combination of interfaces.
Fig. 2 is a system architecture diagram according to an embodiment of the present application. As shown in fig. 2, the system architecture may include a first device 20 and a second device 21.
The first device 20 may be the electronic device 100, and the first device 20 may receive an advertisement request from the second device 21, and may return a recommended search advertisement to the second device 21 according to the advertisement request.
It will be appreciated that in the embodiment of the present application, after receiving the advertisement request of the second device, the first device 20 may determine a corresponding value according to the search content based on the advertisement request, and recommend an advertisement corresponding to the value according to the determined value, and return the ranked search advertisement to the second device 21.
The second device 21 may be an end-side (user-side) device, and the second device 21 may include, but is not limited to, a personal computer (Personal Computer, PC), a mobile phone, a Pad, a tablet, and the like having a display screen, and the type of the second device 21 is not particularly limited in the embodiment of the present application.
It will be appreciated that a target application may be installed on the second device 21, which target application may be used for advertising purposes, which target application may be considered as an intermediary between the user and the advertising platform. The display interface of the target application may include a search box and a search advertisement presentation area. Wherein the search box may be used to input search content for the user and the search advertisement presentation area may be used to present search advertisements returned by the first device. For example, the second device 21 may send an advertisement request to the first device 20 through the target application, receive a search advertisement returned by the first device 20, and display the search advertisement returned by the first device 20 on a display interface of the second device 21.
In some alternative embodiments, the second device 21 may be provided with other applications besides the above-mentioned target application, such as image capturing, image management, message processing, data processing and transmission, text processing, instant message pushing, network communication, media playing, and time management.
It should be noted that, the application scenario of the embodiment of the present application may include, but is not limited to, a search browser or a global search.
Table 1 exemplarily shows the above application scenario and a scenario description thereof.
TABLE 1
Advertising targets for which embodiments of the present application are directed may include, but are not limited to, wake up, stay down, pay, activate, register, form, or other custom user behavior.
Table 2 exemplarily shows specific information of the above-described delivery targets.
TABLE 2
Next, an advertisement recommendation method provided by the embodiment of the present application is illustrated with reference to fig. 3, 4A, 4B, and 5 to 7.
Fig. 3 is a schematic flow chart of an embodiment of an advertisement recommendation method according to the present application, which specifically includes the following steps:
In response to the search content entered by the user, the second device sends an advertisement request to the first device, wherein the advertisement request includes the search content, step 301. Accordingly, the first device receives the advertisement request sent by the second device.
Specifically, when a user has a need to search for advertisements, the target application in the second device may be opened, and search content may be input in a search box of a display page of the target application.
It will be appreciated that the display page of the target application may include a search box in which the user may enter search content. The specific representation of the search box of the display page of the target application may refer to the display interface of the second device in fig. 2, which is not described herein again.
Among other types of advertisements for search advertisements may include, but are not limited to, application advertisements and merchandise advertisements. Application advertising refers to a user returning related applications by searching a content request server. The commodity advertisement refers to a user returning related commodity through the search content request server.
It will be appreciated that the types of search advertisements described above are merely exemplary, and are not limiting of embodiments of the present application, and that in some embodiments, search advertisements may include other types, not just.
In response to a search advertisement entered by a user, the second device may send an advertisement request to the first device, wherein the advertisement request may include the search content.
For example, referring to fig. 4A, taking a search advertisement as an application advertisement as an example, a user may input the following search content: "house source near the south three ring? the user's intent may be to rent or buy a house, thus, the search content may be used to request return to an application associated with the property.
Or referring to fig. 4B, taking the search advertisement as an example of a commercial advertisement, the user may input the following search contents: "is a popular wearing in summer? the user's intent may be to find merchandise related to tidal currents or wearing, thus, the search content may be used to request return of merchandise related to trending or wearing, such as luggage, clothing, gestures, shoes, and the like.
In some alternative embodiments, the advertisement type of the search advertisement may be indicated by the advertisement request. For example, the search box may correspond to an advertisement type of the search advertisement, and after the user inputs the search content in any one of the search boxes, the second device sends an advertisement request to the first device, where the advertisement request carrying the search content may further carry an advertisement type indication, where the advertisement type indication may be used to indicate the advertisement type corresponding to the search box in which the user of the first device currently inputs the search content. Taking the example that the advertisement types of the search advertisement include an application advertisement and a commodity advertisement, the display page of the target application may include 2 search boxes, which may be an application advertisement search box and a commodity advertisement search box, respectively, wherein the application advertisement search box may be used for requesting to return the search advertisement of the application advertisement type, and the commodity advertisement search box may be used for requesting to return the search advertisement of the commodity advertisement type. When the user inputs search content in the application advertisement search box and the second device sends an advertisement request to the first device, the advertisement request may include an advertisement type indication, where the advertisement type indication may indicate that the advertisement type currently requested by the user of the first device is an application advertisement type, or when the user inputs search content in the commodity advertisement search box and the second device sends an advertisement request to the first device, the advertisement request may include an advertisement type indication, where the advertisement type indication may indicate that the advertisement type currently requested by the user of the first device is a commodity advertisement type.
In some alternative embodiments, the advertisement type of the search advertisement may be determined by identifying the search content in the advertisement request. By way of example, the display page of the target application may include 1 search box, which may be used to request a search advertisement to return a different advertisement type. For example, taking an advertisement type of the search advertisement as an example including an application advertisement and a commodity advertisement, when a user inputs search content in a search box and the second device sends an advertisement request to the first device, the first device may identify the search content in the advertisement request to determine the advertisement type of the search advertisement corresponding to the advertisement request, that is, determine whether the advertisement request searches for the application advertisement or the commodity advertisement.
In response to the received advertisement request, the first device determines a target value type corresponding to the advertisement request based on search content in the advertisement request, step 302.
Specifically, after the first device receives the advertisement request, the first device may identify search content in the advertisement request to determine a target value type corresponding to the advertisement request. Wherein, the value type can be used for representing the value of the advertisement request.
By way of example, the value types may include, but are not limited to, low value, which may be used to characterize the value of the current ad request as low, and high value, which may be used to characterize the value of the current ad request as high.
It will be appreciated that the above-described number of types of value types is merely illustrative and not limiting of embodiments of the present application, and in some embodiments, the number of types of value types may be set according to actual requirements, for example, the value types may include low value, medium value, high value, and the like.
The value types, including low value and high value, are now exemplified and described in connection with fig. 5. Referring to fig. 5, the plurality of advertisement requests in the left column contain explicit brand words, e.g.,Such brands, which better express the advertising intentions desired by the user, so that a plurality of advertisement requests in the left column can be regarded as high-value advertisement requests. The content in the plurality of advertisement requests of the right column cannot accurately express the advertisement intention required by the user, and thus the plurality of advertisement requests of the right column may be regarded as low-value advertisement requests.
In some alternative embodiments, the manner in which the first device identifies search content in the advertisement request to determine the value type in the advertisement request may include the first device identifying search content in the advertisement request, obtaining a corresponding keyword, and determining the value type in the advertisement request based on the keyword. For example, the keywords may be used to text match an in-flight advertising listing, which refers to a listing of all advertisements that have been posted on an advertising platform, to determine the value type to which the advertising request corresponds.
Illustratively, text matching is illustratively described as having a value type that includes low value and high value;
if the matching degree of the keyword and the text of the advertisement list is high, the advertisement request value of this time can be considered to be high, namely the value type of the advertisement request of this time is high, or
If the matching degree of the keywords and the text of the advertisement list is low, the advertisement request value of the current time is considered to be low, namely the value type of the advertisement request of the current time is considered to be low.
In some alternative embodiments, the manner in which the value type of the advertisement request is determined may also include counting the frequency of advertisement requests.
For example, if the frequency of advertisement requests for a certain search content is high, the value type of advertisement requests including the search content is high, or if the frequency of advertisement requests for a certain search content is low, the value type of advertisement requests including the search content is low.
In some alternative embodiments, the determination of the value type of the advertisement request may further include displaying revenue by counting the estimated thousands of times the advertisement request (ESTIMATED COST PER MILLE, ECPM).
For example, the advertisement request with high ECPM has a high value or the advertisement request with low ECPM has a low value.
In some alternative embodiments, the manner in which the value type of the advertisement request is determined may also include portrayal by the user.
For example, the number of advertisements clicked by the user within the preset duration may be counted, and if the number of advertisements clicked by the user is greater than or equal to the threshold, the value type of the advertisement request initiated by the user is high value, or if the number of advertisements clicked by the user is less than the threshold, the value type of the advertisement request initiated by the user is low value.
Optionally, step 303 may be performed after step 302:
and step 303, taking the result of determining the target value type corresponding to the advertisement request as a virtual advertisement space.
The result of determining the target value type corresponding to the advertisement request is a virtual ad spot, which includes the determined target value type corresponding to the advertisement request. The virtual ad slots may be presented to an advertiser for viewing on an advertising platform.
More specifically, after determining the target value type corresponding to the advertisement request, compared with the prior art, the advertisement platform can add more information of a virtual advertisement position corresponding to the advertisement request on the display page of the advertisement request, and the virtual advertisement position can display the target value type corresponding to the advertisement request, so that an advertiser can intuitively and specifically know the value of the advertisement request.
Table 3 exemplarily shows virtual ad spot information in a presentation page.
TABLE 3 Table 3
Referring to table 3, the presentation page may include information such as date, real ad spot ID, real ad spot name, virtual ad spot ID, and virtual ad spot name. The virtual advertisement space is an advertisement space invisible to the user, i.e. the virtual advertisement space is not displayed on the second device, but can be displayed on the advertisement platform for the advertiser to view.
Referring next to Table 3, the virtual ad spot information may include a virtual ad spot ID and a virtual ad spot name, and exemplary, real ad spot ID is "k1f9dhqs06" for a virtual ad spot with a high value type and real ad spot ID is "q8oliiiqq5" for a virtual ad spot with a low value type.
Optionally, the first device may also perform step 304 before performing step 302, while performing step 302, or after performing step 302:
In step 304, the first device determines a target advertisement type corresponding to the advertisement request based on the search content in the advertisement request.
The first device may obtain, in addition to the target value type of the current advertisement request, a target advertisement type corresponding to the current advertisement request.
In some alternative embodiments, the advertisement request may carry indication information for indicating the type of advertisement. For example, when the search box corresponds to the advertisement type requested by the user one by one, the advertisement request may carry indication information corresponding to the search box. The advertisement type includes application advertisement and commodity advertisement, if the user sends the advertisement request through the application advertisement search box, the advertisement request can carry indication information indicating that the requested advertisement is the application advertisement, or if the user sends the advertisement request through the commodity advertisement search box, the advertisement request can carry indication information indicating that the requested advertisement is the commodity advertisement.
In some alternative embodiments, the advertisement type may also be determined by identifying the search content in the advertisement request. For example, by identifying the search content in the advertisement request, a user intention corresponding to the advertisement request may be obtained, and advertisement attribute information strongly related thereto may be obtained by the user intention, wherein the advertisement attribute information may be used to determine the type of the search advertisement of the current search.
Next, a manner of determining the keyword and the user intention is exemplified. The search content of the user is "double 11 purchasing strategy recommended instant" and keywords are obtained by identification, and the user intention is shopping by identification. The advertisement attribute information may then be determined by the user intent to be a "procurement" related application.
In step 305, the first device performs advertisement recommendation on the advertisement request based on the target value type corresponding to the advertisement request, so as to obtain N search advertisements corresponding to the target value type.
After the target value type corresponding to the advertisement request is obtained, advertisement recommendation can be performed based on the target value type of the advertisement request so as to obtain N search advertisements.
N is a positive integer, the value of N can be preset according to actual demands, and N search advertisements corresponding to the target value type corresponding to the advertisement request at this time are selected after being arranged according to the ranking order of the search advertisements.
Alternatively, if the first device determines the target value type corresponding to the current advertisement request in step 302 and the first device determines the target advertisement type corresponding to the current advertisement request in step 304, that is, after the target value type and the target advertisement type corresponding to the current advertisement request are obtained, step 305 may not be executed, and instead step 306 may be executed,
In step 306, the first device performs advertisement recommendation on the advertisement request based on the target value type and the target advertisement type corresponding to the advertisement request, so as to obtain N search advertisements corresponding to the target value type and the target advertisement type.
In some alternative embodiments, the N search advertisements may be top N search advertisements selected after being arranged in a ranking order of the search advertisements, where the ranking of the N search advertisements may be determined based on ECPM.
In some alternative embodiments, the value of N may be determined according to the size of the search advertisement presentation area in the second device, where the size information of the search advertisement presentation area in the second device may be carried in the advertisement request, and the form of the search advertisement presentation area may refer specifically to the display interface of the second device in fig. 2, which is not described herein again.
By way of example, assuming a search advertisement presentation area in a second device can accommodate 5 ad slots, a first device, after making an advertisement recommendation based on an advertisement request, can send a top-ranked 5 advertisement to the second device based on the size of the search advertisement presentation area indicated by the advertisement request. Or assuming that the search advertisement presentation area in the second device can accommodate 10 ad spots, the first device, after making an advertisement recommendation based on the advertisement request, may send the top-ranked 10 ads to the second device based on the size of the search advertisement presentation area indicated by the advertisement request.
It is to be appreciated that different value types can correspond to different advertisement recommendation models, wherein different advertisement recommendation models can be utilized to make advertisement recommendations for advertisement requests of different values. For example, taking the example that the value type includes high value and low value, the advertisement recommendation model may include 2 models, the 2 models may be a high value recommendation model and a low value recommendation model, the high value recommendation model may be used to make advertisement recommendation for high value advertisement requests, and the low value recommendation model may be used to make advertisement recommendation for low value advertisement requests, thereby isolating and decoupling advertisement requests of different values to promote recommendation effect of search advertisement.
The first device sends the N search advertisements to the second device, step 307. Accordingly, the second device receives the N search advertisements sent by the first device.
Specifically, after the first device searches for the corresponding N search advertisements based on the advertisement request, the N search advertisements may be transmitted to the second device.
In some alternative embodiments, the first device may also indicate the ranking of the N search advertisements to the second device when the N search advertisements are sent to the second device, e.g., the first device returns the top-ranked N search advertisements.
Fig. 6 illustrates a schematic diagram of a first device returning top-ranked N search advertisements. Referring to fig. 6, the top-N search advertisement may be a top-N advertisement selected by sorting according to ECPM, for example, the top-N search advertisement may include app_1, app_2, and app_n, where app_1 is highest in rank, app_2 is second highest in rank, and so on.
The second device displays the N search advertisements, step 308.
Specifically, after the second device receives the N search advertisements sent by the first device, the N search advertisements may be displayed in a search advertisement presentation area of the second device.
In some alternative embodiments, after the second device receives the top-N search advertisement sent by the first device, the top-N search advertisement may be displayed in the search advertisement presentation area of the second device in the ranking order.
Fig. 7 illustrates a schematic diagram of a second device displaying top-ranked N search advertisements. Referring to fig. 7, after receiving the top-ranked N search advertisement returned by the first device, the second device may sequentially display the top-ranked N search advertisement in the search advertisement presentation interface in a top-down order, where the top-ranked search advertisement is ranked higher. Illustratively, assuming that ECPM of app_1 is highest, ranked first, ECPM times of APP __2, ranked second, ECPM of app_n is lowest, and ranked last, app_1 may be displayed uppermost in the search advertisement presentation interface, app_2 may be displayed below app_1, and app_n may be displayed lowermost in the search advertisement presentation interface.
It should be understood that the manner in which the ranking levels of the search advertisements are sequentially displayed in the order from top to bottom is merely illustrative, and not limiting to the embodiment of the present application, it is merely one display manner according to the visual habit of the user, and in some embodiments, other display manners may be provided, for example, the ranking levels of the search advertisements may be sequentially displayed in the order from left to right.
In the existing technical scheme of advertisement recommendation, advertisement recommendation is carried out on advertisement requests through a unified model, namely advertisement recommendation is carried out through a unified recommendation algorithm, the recommendation result does not distinguish the value of the advertisement requests, the advertisement recommendation result and the intention of a user are easy to match, and the advertisement recommendation effect is poor.
The application identifies the corresponding value of the advertisement request and carries out adaptive advertisement recommendation according to the value to obtain the corresponding search advertisement, compared with the prior art, the application does not distinguish the value of the advertisement request, according to the application, advertisement requests with different values can be isolated and decoupled, so that the recommendation effect of the search advertisement is improved, the recommended search advertisement is more in line with the intention of a user, and the satisfaction degree of the user and an advertiser is improved.
It can be understood that, search advertisements obtained by recommending through recommendation models corresponding to different value types can be placed in corresponding virtual advertisement positions, for example, search advertisements obtained by recommending advertisements through recommendation models corresponding to high-value advertisement types are placed in corresponding high-value virtual advertisement positions, and search advertisements obtained by recommending advertisements through recommendation models corresponding to low-value advertisement types are placed in corresponding low-value virtual advertisement positions. While these virtual ad slots are not visible to the end user, these virtual ad slots may be presented to the advertiser. On the basis, the advertisement platform can set different bidding strategies for different virtual advertisement positions so as to improve the competitiveness of advertisements and the benefit of the advertisement platform.
The advertisement recommendation method is exemplified by fig. 3, 4A, 4B, and 5-7. Next, an advertisement recommendation method is exemplarily described below by way of specific embodiments in conjunction with fig. 8 and 9.
Fig. 8 is a schematic system architecture diagram of a first device according to an embodiment of the present application. Referring to fig. 8, the system architecture of the first device may include a search content identification unit, a value processing unit, and an advertisement recommendation unit.
The search content identification unit may be configured to identify the search content in the advertisement request by means of natural language processing (Natural Language Understanding, NLU) or the like, so as to obtain keywords in the search content.
In some alternative embodiments, the search content identification unit may be further configured to identify search content in the advertisement request to determine an advertisement type corresponding to the advertisement request.
The value processing unit may be configured to determine a value type corresponding to the advertisement request based on the keyword. The value types corresponding to the advertisement request can be classified into a high value type, a medium value type and a low value type, and can be distinguished according to the rest value rules.
It should be noted that the result of determining the value type corresponding to the advertisement request is a virtual ad slot, i.e. the virtual ad slot includes the value type corresponding to the advertisement request already determined. The virtual ad slots may be presented to an advertiser for viewing on an advertising platform.
The advertisement recommendation unit may be configured to perform corresponding advertisement recommendation based on the value type to obtain a search advertisement corresponding to the value type.
In some alternative embodiments, the advertisement recommendation unit may be implemented by a recommendation algorithm of a Demand-Side Platform (DSP).
Wherein, the DSP can comprise a recall subunit, a coarse arranging subunit and a fine arranging subunit. The recall subunit may be configured to recall the advertisement to be selected via a recall algorithm corresponding to the value type. The coarse ranking subunit may be configured to coarse rank the recalled candidate advertisements by a coarse ranking algorithm corresponding to the value type. The fine ranking subunit may be configured to fine rank the coarse ranked advertisements by a fine ranking algorithm corresponding to the value type to obtain fine ranked advertisements, where the fine ranked advertisements may be top-ranked N search advertisements, and the top-ranked N search advertisements may be used for returning to the second device.
In some alternative embodiments, the first device may further forward the advertisement request to the third party DSP, and may fuse the refined results with the output results of the third party DSP to obtain the top-ranked N search advertisement.
It may be understood that the DSP corresponding to the advertisement recommendation unit is a DSP of an advertisement platform to which the first device belongs, and the third party DSP may be a DSP of a third party advertisement platform.
Next, on the basis of fig. 8, the manner of generating the search advertisement in the first device is exemplarily described below with reference to fig. 9.
As shown in fig. 9, which is a flowchart illustrating another embodiment of the advertisement recommendation method provided by the present application, the steps 302 and 305 may specifically include the following steps:
In step 901, the first device identifies search content in the advertisement request to obtain keywords.
Specifically, after receiving the advertisement request, the first device may identify the search content in the advertisement request by means of NLU or the like, so as to obtain a keyword corresponding to the search content.
The keyword obtaining manner may specifically refer to the related description in the foregoing embodiments, which is not described herein again.
In step 902, the first device determines a target value type corresponding to the advertisement request based on the keyword.
Specifically, after the first device obtains the keyword, the target value type corresponding to the advertisement request may be determined based on the keyword.
The method for determining the target value type of the advertisement request can be realized by text matching of keywords and the advertisement listing. The text may be matched in such a way that the keywords are matched to the names of all advertisements in the list of advertisements.
Illustratively, an on-air advertising list may be acquired first. Wherein, the advertisement casting list can be obtained by counting all advertisement casting in the advertisement platform.
In some alternative embodiments, since the in-flight advertisement on the advertisement platform is dynamically updated, the in-flight advertisement on the advertisement platform may be monitored for a preset period of time, and when the in-flight advertisement on the advertisement platform changes, the in-flight advertisement list may be updated.
Then, after the advertisement in progress list is obtained, the keywords can be matched with the advertisement in progress list in text so as to obtain the value type corresponding to the advertisement request.
In some alternative embodiments, the manner in which keywords are text matched with the listing may be based on a brand word policy, for example, brand words in the keywords may be obtained and the matching may be based on the brand words with text in the listing.
If the brand word in the keyword is matched with any advertisement in the advertisement list and the matching degree is higher, the value of the advertisement request is higher, or
If the brand word in the keyword is matched with any advertisement in the advertisement casting list and the matching degree is lower, the value of the advertisement request is lower.
In some alternative embodiments, the results of the keyword matching the text in the advertisement listing may include a plurality of results, for example, the results of the text matching may include a plurality of results of no match, exact match, prefix match, and inclusive match.
Table 4 exemplarily shows the results of various text matches.
TABLE 4 Table 4
| Searching content | Text matching results | Matched advertisements |
| "How weather today" | Mismatch of | Without any means for |
| 'Jiaduoduo@' | Accurate matching | Multiple pieces@ |
| 'Spell' | Prefix matching | Multiple pieces@ |
| "How to view the mosaic of multiple@ streams" | Containing matches | Multiple pieces@ |
Referring to table 4, for the search content of "how today's weather" no advertisement of any brand is matched, and thus, the result of text matching is "no match". For the search content of "multiple of pieces@", an advertisement with a brand of "multiple of pieces@" is matched, and thus, the result of text matching is "exact matching". For "spelled" search content, the search content matches the prefix of the brand "how much@" is spelled, and therefore, the result of the text match is "prefix match". For the search content of "how to view the multi-part@ stream", the search content contains the brand "multi-part@", and thus, the result of the text matching is "contain matching".
It will be appreciated that the above-described plurality of text matching results are merely illustrative and not limiting of embodiments of the present application, and in some embodiments, more or fewer text matching results than those described above may be included.
In some alternative embodiments, the result of the text matching may form a mapping relationship with the value type, the mapping relationship may be one-to-one, or the mapping relationship may be many-to-one, which is not particularly limited by the embodiment of the present application.
By way of example, taking the case that the text matching result includes a plurality of results such as mismatch, exact match, prefix match, and include match, and the mapping relationship between the text matching result and the value type is a many-to-one example, table 4 exemplarily shows the mapping relationship between the text matching result and the value type.
TABLE 5
Referring to table 5, the result of the text match of "no match" may be mapped with a low value type, and the result of the text match of "exact match", "prefix match" and "contain match" may be mapped with a high value type.
It will be appreciated that the above-mentioned mapping relationship between the text matching result and the value type is merely illustrative, and not limiting to the embodiments of the present application, and in some embodiments, the mapping relationship between the text matching result and the value type may be set according to actual requirements.
It should be noted that, in some embodiments, the first device determines the value type of the advertisement request based on the keyword as a virtual advertisement slot. The virtual ad slots include the value types to which the ad requests have been determined. The virtual ad slots may be presented to the advertiser for viewing on the advertising platform so that the advertiser intuitively perceives the value of the ad placement.
In step 903, the first device performs advertisement recommendation on the advertisement request based on the target value type, and obtains N search advertisements corresponding to the target value type.
Specifically, after obtaining the target value type corresponding to the advertisement request, the first device may recommend advertisement to the advertisement request based on the target value type, so as to obtain N search advertisements corresponding to the target value type.
It will be appreciated that different recommendation models may be used to recommend advertisement requests of different value types for different value types.
For example, taking the example that the value type includes high value and low value, if the value type of the current advertisement request is low value, a search may be performed using a recommendation model corresponding to the low value to obtain a recommended advertisement corresponding to the low value, or if the value type of the current advertisement request is high value, a recommendation may be performed using a recommendation model corresponding to the high value to obtain a search advertisement corresponding to the high value.
In some alternative embodiments, the recommendation model may be obtained after pre-training, e.g., still taking the example that the value type includes high value and low value, the recommendation model corresponding to low value may be trained using sample data containing low value, or the recommendation model corresponding to high value may be trained using sample data containing high value.
In some alternative embodiments, the recommendation model may include a recall sub-model, a coarse ranking sub-model, and a fine ranking sub-model, wherein at least one of the recall sub-model, the coarse ranking sub-model, and the fine ranking sub-model may use a model corresponding to the value type of the current advertisement request.
For example, a recall sub-model may be used that corresponds to the value type of the current advertisement request, wherein the recall sub-model may be obtained in advance after training with sample data that includes low and high values, and a coarse ranking sub-model and a fine ranking sub-model may be used that is not trained with sample data that includes low and high values.
For another example, a recall sub-model and a coarse ranking sub-model may be used that correspond to the value type of the current advertisement request, wherein the recall sub-model and the coarse ranking sub-model may be obtained in advance after training with sample data that includes low value and high value, and a fine ranking sub-model may be used that is not trained with sample data that includes low value and high value.
For another example, a recall sub-model, a coarse ranking sub-model, and a fine ranking sub-model may be used that correspond to the value type of the current advertisement request, where the recall sub-model, coarse ranking sub-model, and fine ranking sub-model may be obtained in advance after training by sample data that includes low value and high value.
Therefore, different advertisement requests can be isolated and decoupled through different value types, and the advertisement requests with different value types are recommended through different recommendation models, so that search advertisements corresponding to the different value types can be obtained, the recommendation effect can be improved, and poor recommendation effect caused by processing advertisement requests with different values through a unified model is avoided.
Next, taking a case where the recommendation model includes a recall sub-model, a coarse ranking sub-model, and a fine ranking sub-model as an example, an acquisition process of the search advertisement will be exemplarily described. Assuming that the value type of the current advertisement request is high-value, the recommendation can be performed through a high-value recall sub-model, a high-value coarse ranking sub-model and a high-value fine ranking sub-model. First, the first device may recall the in-flight advertisement through a high value recall sub-model to obtain a recall advertisement. The first device may then coarse-rank filter the recall advertisements via the high-value coarse-rank sub-model to obtain coarse-rank advertisements. And then carrying out fine-ranking filtration on the recall advertisements through a high-value fine-ranking sub-model so as to obtain fine-ranking advertisements.
It will be appreciated that each of the fine-ranked advertisements has a corresponding predicted click-Through Rate (pCTR) and predicted conversion Rate (PREDICTED CONVERSION RATE, pCVR).
Finally, the fine-ranked advertisements can be ranked through ECPM, and the advertisement with the top N rank can be obtained from the ranked fine-ranked advertisements to serve as a search advertisement.
Wherein ECPM of each advertisement in the fine-ranked advertisements can be obtained by calculation according to the following formula:
ECPM=pCTR×pCVR×bid×pid;
where bid is the bid of the advertisement and pid is the system bid factor.
By way of example, assuming 1200 advertisements are recalled by the high value recall sub-model, then 200 coarse-ranked advertisements may be obtained after coarse-ranked filtering of the 1200 recalled advertisements by the high value coarse-ranked sub-model, and then the ECPM of the 200 coarse-ranked advertisements may be calculated by the high value fine-ranked sub-model, 200 fine-ranked advertisements may be obtained. Finally, the 200 fine-ranked advertisements are ranked according to ECPM, and the top-ranked N advertisements can be obtained as search advertisements.
In some optional embodiments, the N search advertisements ranked N before may be further obtained after ranking by a first candidate search advertisement and a second candidate search advertisement, where the first candidate search advertisement may be a candidate advertisement obtained after the advertisement recommendation by the DSP through the advertisement recommendation unit, and the second candidate search advertisement may be a candidate advertisement obtained after the advertisement recommendation by the third party DSP.
For example, the first candidate search advertisement and the second candidate search advertisement may be fused to form a set of candidate advertisements, where each advertisement in the set of candidate advertisements has a corresponding ECPM. Then, ranking all advertisements in the advertisement set to be selected according to ECPM of each advertisement, and obtaining the advertisement with the top N rank, wherein the advertisement with the top N rank is used as the search advertisement requested by the user.
It may be appreciated that the second search advertisement to be selected may be obtained after advertisement recommendation by the third party DSP through its own recommendation algorithm, or the second search advertisement to be selected may be obtained after advertisement recommendation by the third party DSP through an existing or disclosed recommendation algorithm, which is not particularly limited in the embodiment of the present application.
Fig. 10 is a schematic structural diagram of an embodiment of an advertisement recommendation apparatus according to the present application, as shown in fig. 10, where the advertisement recommendation apparatus 1000 is applied to a first device, the advertisement recommendation apparatus 1000 may include a receiving module 1010, a determining module 1020, a recommending module 1030, and a transmitting module 1040, where,
A receiving module 1010, configured to receive an advertisement request sent by a second device, where the advertisement request includes search content;
a determining module 1020, configured to determine a target value type corresponding to the advertisement request based on the search content in the advertisement request;
A recommending module 1030, configured to recommend an advertisement to the advertisement request based on a target value type corresponding to the advertisement request, and obtain N search advertisements corresponding to the target value type, where N is a positive integer;
and a sending module 1040, configured to send the N search advertisements to the second device.
In one possible implementation manner, the determining module 1020 is further configured to use a result of determining the target value type corresponding to the advertisement request as the virtual advertisement slot.
In one possible implementation manner, the recommending module 1030 is further configured to determine a target advertisement type corresponding to the advertisement request based on the search content in the advertisement request;
And recommending the advertisement request based on the target value type and the target advertisement type corresponding to the advertisement request, and obtaining N search advertisements corresponding to the target value type and the target advertisement type.
In one possible implementation manner, the recommendation module 1030 is further configured to determine a user intention based on the search content in the advertisement request;
and determining a target advertisement type corresponding to the advertisement request based on the user intention.
In one possible implementation, the advertisement types include at least application advertisements and merchandise advertisements.
In one possible implementation, the N search advertisements are top-ranked N search advertisements, and the ranking of the N search advertisements is determined by estimating thousands of presentation benefits ECPM.
In one possible implementation manner, the N search advertisements are obtained after ranking based on a first search advertisement and a second search advertisement, where the first search advertisement is obtained after advertisement recommendation in the first device, and the second search advertisement is obtained after advertisement recommendation in a third device, and the third device is a device of a third party requester platform DSP.
In one possible implementation manner, the recommendation module 1030 is further configured to determine a target recommendation model corresponding to the target value type based on the target value type corresponding to the advertisement request;
And recommending the advertisement request by using the target recommendation model corresponding to the target value type to obtain N search advertisements corresponding to the target value type.
In one possible implementation manner, the determining module 1020 is further configured to identify the search content in the advertisement request, and obtain a keyword;
And carrying out text matching on the keywords and the advertisement casting list, and determining the target value type corresponding to the advertisement request.
In one possible implementation manner, the determining module 1020 is further configured to periodically update the on-air advertisement list.
In one possible implementation, the text match includes a plurality of results;
the multiple results and multiple value types are in one-to-one mapping relation, or
The multiple results and the multiple value types are in a mapping relation of many to one.
In one possible implementation, the advertisement recommendation apparatus 1000 may be a chip or a first device.
The advertisement recommendation apparatus 1000 provided in the embodiment shown in fig. 10 may be used to implement the technical solution of the method embodiment of the present application, and the implementation principle and technical effects may be further described with reference to the related description in the method embodiment.
It should be understood that the above division of the modules of the advertisement recommendation apparatus 1000 shown in fig. 10 is merely a division of a logic function, and may be fully or partially integrated into a physical entity or may be physically separated. The modules can be realized in the form of software calling through the processing element, can be realized in the form of hardware, can also be realized in the form of software calling through the processing element, and can be realized in the form of hardware. For example, the detection module may be a separately established processing element or may be implemented integrated in a certain chip of the electronic device. The implementation of the other modules is similar. In addition, all or part of the modules can be integrated together or can be independently implemented. In implementation, each step of the above method or each module above may be implemented by an integrated logic circuit of hardware in a processor element or an instruction in a software form.
For example, the modules above may be one or more integrated circuits configured to implement the methods above, such as one or more Application SPECIFIC INTEGRATED Circuits (ASIC), or one or more microprocessors (DIGITAL SIGNAL Processor; DSP), or one or more field programmable gate arrays (Field Programmable GATE ARRAY; FPGA), or the like. For another example, the modules may be integrated together and implemented in the form of a System-On-a-Chip (SOC).
In the above embodiments, the processor may include, for example, a CPU, a DSP, a microcontroller, or a digital signal processor, and may further include a GPU, an embedded neural network processor (Neural-network Process Units; hereinafter referred to as NPU), and an image signal processor (IMAGE SIGNAL Processing; hereinafter referred to as ISP), where the processor may further include a necessary hardware accelerator or a logic Processing hardware circuit, such as an ASIC, or one or more integrated circuits for controlling the execution of the program according to the technical solution of the present application. Further, the processor may have a function of operating one or more software programs, which may be stored in a storage medium.
The embodiment of the application also provides a readable storage medium, wherein a program is stored in the readable storage medium, and when the readable storage medium runs on the electronic device, the electronic device is caused to execute the method provided by the embodiment of the application.
The embodiments of the present application also provide a program product comprising a program which, when run on an electronic device, causes the electronic device to perform the method provided by the illustrated embodiments of the present application.
In the embodiments of the present application, "at least one" means one or more, and "a plurality" means two or more. "and/or", describes an association relation of association objects, and indicates that there may be three kinds of relations, for example, a and/or B, and may indicate that a alone exists, a and B together, and B alone exists. Wherein A, B may be singular or plural. The character "/" generally indicates that the context-dependent object is an "or" relationship. "at least one of the following" and the like means any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c may represent a, b, c, a and b, a and c, b and c, or a and b and c, wherein a, b, c may be single or plural.
Those of ordinary skill in the art will appreciate that the various elements and algorithm steps described in the embodiments disclosed herein can be implemented as a combination of electronic hardware, computer software, and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In several embodiments provided by the present application, any of the functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. The storage medium includes various media capable of storing program codes, such as a U disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk.
The foregoing is merely exemplary embodiments of the present application, and any person skilled in the art may easily conceive of changes or substitutions within the technical scope of the present application, which should be covered by the present application. The protection scope of the present application shall be subject to the protection scope of the claims.