Business object recommendation method, device, electronic device and storage medium

Abstract

The present application discloses a method, device, electronic device and storage medium for recommending a business object, the method comprising: obtaining at least one candidate business object under a target business, each candidate business object corresponding to at least one target tag; obtaining at least one reference tag generated by a business user of the target business in a reference business, wherein the reference business is different from the target business, and the reference tag is determined according to the attention behavior of the business user in the reference business; determining a recommendation score for each candidate business object according to the at least one reference tag and the at least one target tag; determining a target business object from the at least one candidate business object according to the recommendation score and recommending it to the business user. The present application can realize accurate business object recommendation.

Description

Technical Field

[0001] The present application relates to the field of information processing technology, and more specifically, to a business object recommendation method, device, electronic device and storage medium. Background Art

[0002] With the development of Internet technology, more and more business types are provided by business systems. Among them, the attention type business is a common business type, which can provide users with a content attention function so that users can use this function to browse the content they are interested in again. At present, business systems can also use this function to recommend relevant content to users, so as to make targeted content recommendations. However, the current recommendation method is single and still has the problem of low accuracy. Summary of the invention

[0003] The embodiments of the present application provide a business object recommendation method, device, electronic device and storage medium, which can improve the accuracy of business object recommendation.

[0004] In a first aspect, an embodiment of the present application provides a business object recommendation method, the method comprising: obtaining at least one candidate business object under a target business, each of the candidate business object corresponding to at least one target tag; obtaining at least one reference tag generated by a business user of the target business in a reference business, wherein the reference business is different from the target business, and the reference tag is determined based on the attention behavior of the business user in the reference business; determining a recommendation score for each of the candidate business objects based on the at least one reference tag and the at least one target tag; and determining a target business object from the at least one candidate business object based on the recommendation score and recommending it to the business user.

[0005] In a second aspect, an embodiment of the present application provides a business object recommendation device, which includes: a candidate object acquisition module, used to acquire at least one candidate business object under a target business, each of the candidate business object corresponds to at least one target tag; a reference tag acquisition module, used to acquire at least one reference tag generated by a business user of the target business in a reference business, wherein the reference business is different from the target business, and the reference tag is determined based on the attention behavior of the business user in the reference business; a recommendation score evaluation module, used to determine the recommendation score of each of the candidate business objects based on the at least one reference tag and the at least one target tag; a target object recommendation module, used to determine a target business object from the at least one candidate business object based on the recommendation score and recommend it to the business user.

[0006] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory; one or more processors coupled to the memory; one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the business object recommendation method provided in the first aspect above.

[0007] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a program code is stored. The program code can be called by a processor to execute the business object recommendation method provided in the first aspect above.

[0008] The embodiments of the present application provide a method, device, electronic device and storage medium for recommending a business object, by obtaining at least one candidate business object under the target business, wherein each candidate business object corresponds to at least one target tag, and obtaining at least one reference tag generated by the business user of the target business in the reference business, wherein the reference business is different from the target business, and the reference tag is determined according to the attention behavior of the business user in the reference business, and then the recommendation score of each candidate business object is determined according to the at least one reference tag and the at least one target tag, so as to determine and recommend the target business object recommended to the business user from the at least one candidate business object according to the recommendation score. The present application can classify the interests and hobbies of the user under different business types by labeling the attention behavior of the user under different business types, so that when recommending the business object under the target business to the business user under the target business, it can comprehensively evaluate the acceptance of the business user to each candidate business object according to the interests and hobbies of the business user under other business types, so that according to the evaluation result, the business object for the business user can be accurately recommended. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 An application scenario diagram of a business object recommendation method provided in an embodiment of the present application is shown.

[0011] Figure 2 A business process diagram provided in an embodiment of the present application is shown.

[0012] Figure 3 A schematic diagram of an interface for storing attention data provided in an embodiment of the present application is shown.

[0013] Figure 4 Another business process diagram provided by an embodiment of the present application is shown.

[0014] Figure 5 A flow chart of a business object recommendation method provided in one embodiment of the present application is shown.

[0015] Figure 6 A flow chart of a business object recommendation method provided in another embodiment of the present application is shown.

[0016] Figure 7 A flow chart of a business object recommendation method provided in yet another embodiment of the present application is shown.

[0017] Figure 8 A flow chart of a business object recommendation method provided in yet another embodiment of the present application is shown.

[0018] Fig. 9 A flow chart of step S430 in a business object recommendation method according to another embodiment of the present application is shown.

[0019] Fig.10 A flow chart of step S432 in a business object recommendation method according to another embodiment of the present application is shown.

[0020] Fig.11 A flowchart of a business object recommendation method provided by another embodiment of the present application is shown.

[0021] Fig.12 A module block diagram of a business object recommendation device provided in an embodiment of the present application is shown.

[0022] Fig.13 A structural block diagram of an electronic device provided in an embodiment of the present application is shown.

[0023] Fig.14 A storage unit for storing or carrying program codes for implementing the business object recommendation method according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0025] See also Figure 1 , Figure 1A schematic diagram of an application scenario of the business object recommendation method provided in an embodiment of the present application is shown, and the application scenario includes a business system 10 provided in an embodiment of the present application. The business system 10 includes: a terminal device 100 and a server 200. Among them, the terminal device 100 and the server 200 are located in a wireless network or a wired network, and the terminal device 100 and the server 200 can interact with data. In some embodiments, there can be multiple terminal devices 100, and the server 200 can be communicatively connected with multiple terminal devices 100. Multiple terminal devices 100 can also be communicatively connected to each other through the Internet, or the server 200 can be used as a transmission intermediary to achieve data interaction between each other through the Internet.

[0026] In the embodiment of the present application, the terminal device 100 may be a mobile phone, a smart phone, a laptop computer, a desktop computer, a tablet computer, a personal digital assistant (PDA), a media player, a smart TV, a wearable electronic device, etc. The specific terminal device type may not be limited in the embodiment of the present application. The server 200 may be a separate server or a server cluster, a local server or a cloud server, and the specific server type may not be limited in the embodiment of the present application.

[0027] In some embodiments, a client may be installed in the terminal device 100. The client may be a computer application (Application, APP) installed on the terminal device 100, or a Web client, which may refer to an application developed based on a Web architecture. When a user logs in to the client through an account, all information corresponding to the account may be stored in the storage space of the server 200. The information corresponding to the account includes information input by the user through the client and information received by the user through the client.

[0028] In some embodiments, the client may be an application for providing content services, such as a novel reading platform, an e-book payment platform, a video platform, a game platform, and other applications. In some embodiments, the client may provide a follow-up service, so that the user can use the client to follow the content of interest. In some embodiments, the service system may store the user's follow-up behavior data to the server, and may also send relevant notifications or recommend relevant content to the client used by the user based on the follow-up behavior data.

[0029] For example, on a novel reading app, when a user follows a novel author, after the novel author publishes a new novel, the business system can send a notification to the user based on the follow relationship to recommend the novel author's new book. For another example, if a user "likes" a chapter or votes for a novel, the business system can send a novel ranking list to users who have "liked" or voted, and recommend the books on the list. For another example, on a book e-commerce website, if a user collects a book, when the price of the book changes or there is a discount on books of the same type, the website can recommend it to the user.

[0030] For example, see Figure 2 , Figure 2 A business process diagram provided by an embodiment of the present application is shown. It can be seen that the specific business processes of the above-mentioned attention-type businesses can be summarized as follows: as a follower (i.e., the initiator of the attention behavior in the business), the user can initiate a request to the business system to establish a follow-up relationship with a certain business object, and the business system can record the user's attention relationship data; the follower can query his own follow-up list, or query whether he has already followed a certain business object, and can also further obtain the content services provided by the followed object (i.e., the recipient of the attention behavior in the business) through the follow-up list; the followed object can query which users follow him (i.e., the fan list); when the followed object publishes new content or other changes, the business system can notify the corresponding followers according to its fan list, and can also recommend other business objects to the followers according to the characteristics of the followed object and provide new content services. Among them, attention-type businesses such as attention, subscription, voting, likes, and collections all conform to the above-mentioned process.

[0031] After research, the inventors found that there is a problem of low development efficiency when developing these attention-type businesses. Specifically, if each business scenario develops its own relationship storage service, then there will be multiple sets of different implementations of homogeneous services in the same system. For example, an interface for determining whether a user has established a follow-up relationship needs to be provided separately for each business, and the reuse rate is relatively low. Moreover, when certain operational activities need to be launched quickly, if a follow-up relationship is required, the workload of developing the storage service must be added to the workload assessment, which greatly reduces the development efficiency. Furthermore, because each business scenario develops its own relationship storage service, the attention behavior data between different businesses are not connected, so that the recommended data of each business scenario cannot refer to the attention behavior data of other businesses, resulting in low accuracy of the recommended data.

[0032] In the embodiments of the present application, in response to the above-mentioned problem of low development efficiency, the inventors, after long-term research, proposed a universal attention system provided in the embodiments of the present application, so that the business system does not need to implement attention data storage by itself, but the attention data storage is uniformly taken over by the universal attention system, thereby improving business development efficiency.

[0033] In some embodiments, the above-mentioned unified takeover of attention data storage can be the storage of forward relationship data and reverse relationship data. Specifically, for attention type services, it is actually a mapping relationship established between the user and the business object. From the perspective of the follower, the mapping relationship is [follower->followed object], which is the forward relationship data; from the perspective of the followed object, the mapping relationship is [followed object->follower], which is the reverse relationship data. Since the logic of the mapping relationship is the same for different attention type services, the above-mentioned relationship data can be taken over and stored through the universal attention system of the present application. In some embodiments, the forward relationship data and the reverse relationship data can be checked and synchronized regularly to maintain the final consistency of the data.

[0034] In other embodiments, the above-mentioned unified takeover of follow data storage may also be only the storage of forward relationship data, that is, the forward relationship data is the basic data that must be stored, and the reverse relationship data is the data that can be optionally stored. It can be understood that the storage of reverse relationship data is mainly to facilitate the acquisition of the follower list of the followed object. However, since some business scenarios do not require the provision of a follower list, while some business scenarios do require it, it can be determined whether to store reverse relationship data based on the actual business scenario, which is not limited here. For example, some business scenarios may have certain hot business objects, which causes the amount of data in its follower list to increase dramatically, resulting in greater data access pressure. The data access pressure can be dispersed by providing reverse relationship data storage for them.

[0035] Because when the general attention system stores attention data, it is not necessary to know the meaning of different followers or followed objects in specific business scenarios, nor is it necessary to store their complete business information. Therefore, in some embodiments, when storing attention data, for different followers or followed objects, the identifier (Identity Document, ID) of the follower or followed object can be used as the storage basis. Exemplarily, when storing the attention relationship data of a certain user, the storage format of the basic data (i.e., the forward relationship data) can be [Follower Identification ID: Followed Object Identification ID]. It can be understood that when it is necessary to store reverse relationship data, its storage format can be [Followed Object Identification ID: Follower Identification ID].

[0036] For example, in the book following service in a novel reading app, the followers are the reading users and the followed objects are the books. When a user follows a novel, the information stored in the general following system is the user ID and the book ID, and it is not necessary to store the complete user information and book information.

[0037] In this way, even if multiple businesses are connected to the universal attention system of this application at the same time, the amount of stored data will not grow too fast, reducing storage costs. There is no need to adjust the storage format of the underlying data (fields in the database table), which greatly reduces the complexity of data storage and development costs.

[0038] In some embodiments, in order to ensure that different concerned services do not affect each other, a unique service identification ID may be assigned to each concerned service. For example, when storing the concerned relationship data of a user of a service, the basic storage format may be [service identification ID: follower identification ID: concerned object identification ID].

[0039] In some embodiments, the storage resource instance implementation may be configurable so that each business may choose to allocate an independent storage database instance. The type of the database may be MySQL / MongoDB / KV, etc., which is not limited here. The database types corresponding to different businesses may be the same or different, which is not limited here.

[0040] It is understandable that after different services are connected, independent business identification IDs and corresponding storage resources can be allocated. Therefore, even if a business system or storage resource fails, it will not affect the normal operation of other businesses. You only need to troubleshoot and restore the problem business.

[0041] For example, see Figure 3 , Figure 3 Specifically, when the concerned business needs to access the general system of this application, relevant business information can be filled in to apply for business identification and storage resources, such as Figure 3 The business identification ID, database type, and database name corresponding to the different business names shown. Optionally, you can also fill in whether to store reverse relationship data. If so, the subsequent general attention system will automatically record the reverse relationship data when storing attention data.

[0042] In some embodiments, a universal interface may be provided for various attention-type business scenarios in a unified manner, so that various execution logics of attention relationships may be implemented by calling the universal interface, including adding / deleting attention relationships, querying whether an attention relationship exists, querying the list and number of objects followed by a certain follower, querying the list and number of followers of a certain followed object, etc. Among them, the basic parameters of the universal interface may be at least one of the business identification ID, the follower identification ID, and the followed identification ID. In some embodiments, different execution logics may correspond to different universal interfaces, so that when the target execution logic needs to be implemented, the target universal interface corresponding to the target execution logic may be called.

[0043] Specifically, adding / deleting follow-up relationship data can be achieved by modifying the forward relationship data; querying whether a follow-up relationship exists can be achieved by querying the forward relationship data; querying the list and number of followed objects of a certain follower can be achieved by querying the forward relationship data; querying the list and number of followers of a certain followed object can be achieved by querying the reverse relationship data.

[0044] In some embodiments, if the business system needs to provide services in different data centers (computer rooms), the general interest interface service instances and storage resources can be flexibly deployed nearby to reduce interface latency.

[0045] For example, see Figure 4 , through the universal attention system of the present application, the process of a attention behavior can be: after the business system is connected to the universal attention system of the present application, the follower initiates a request and sends the request content to the specific business system. After the business system receives the request, it can call the corresponding universal interface of the universal attention system according to the request type. The universal attention system receives the request and finds the corresponding storage database instance according to the parameters in the request (business identifier, follower identifier, followed person identifier), performs data modification or data query, and after the processing is completed, the reply content can be sent to the business system. The business system receives the reply, can process the reply content, and return the final result to the follower. In some embodiments, when the reply received by the business system includes the relevant object identifier, the business system can fill in the identification information corresponding to the relevant object identifier according to its own needs to display the complete content to the user.

[0046] In the embodiment of the present application, in order to solve the problem of low recommendation accuracy, the embodiment of the present application further provides a business object recommendation method, device, system, electronic device and storage medium based on the above-mentioned general attention system, which can improve the accuracy of business object recommendation. The following will be described in detail through specific embodiments.

[0047] See also Figure 5 , Figure 5 A flow chart of a business object recommendation method provided by an embodiment of the present application is shown, which can be applied to an electronic device, wherein the electronic device can be a server corresponding to the above-mentioned general attention system, or a server corresponding to the above-mentioned business system, or the above-mentioned terminal device, which is not limited here. The business object recommendation method may include:

[0048] Step S110: obtaining at least one candidate business object under the target business, each of the candidate business objects corresponding to at least one target tag.

[0049] Since the attention behavior data between different businesses are not connected, the recommendation data of a certain business is determined only by the attention behavior data of the current business. However, the user's attention behavior in the current business may not fully reflect the user's actual interest preferences, so if the business object is recommended to the user only based on the attention behavior of the current business, the recommendation accuracy may not be high. Therefore, in an embodiment of the present application, based on the above-mentioned general attention system, the user's interest preferences reflected in other businesses can be determined, and then the user's interest preferences between different businesses can be combined to accurately recommend business objects under the current business to the user.

[0050] Among them, the business object can be an object related to the service content provided by the business. Among them, the business is a collection of services provided to users, which can be online video services, online audio services, and online text services. Videos can be movies, TV series, or user-made videos. Audio can be popular songs, music, or user-made audio. Texts such as novels, reference books, or professional dictionaries. As a way, the business object can be the specific service content provided by the business. For example, a novel, a video, a piece of music, etc. As another way, the business object can also be a person related to the service content, which can be the creator of the service content or the participant of the service content. For example, the author of a novel, the original creator of a homemade video, the actor or director in a video, the singer of a piece of music, etc. As another way, the business object can also be a business user, that is, the user of the business. For example, in the novel reading APP, reading user A, reading user B, etc. In some embodiments, the business user can also be understood as the terminal used by the user.

[0051] In an embodiment of the present application, when a business object under the current business needs to be recommended to the user, the electronic device can obtain at least one candidate business object under the target business. The target business can be a business that needs to be recommended, and the candidate business object is an object related to the service content provided by the target business, which can be a business object that needs to be evaluated for recommendation.

[0052] In some embodiments, the candidate business object can be a business object with a certain recommendation probability in the target business for the business user, so that further recommendation evaluation can be made for the candidate business object by combining the user's interest preferences reflected in other businesses, thereby filtering out inaccurately recommended candidate business objects and achieving accurate recommendation of business objects.

[0053] As a method, the electronic device can obtain the hot business objects under the target business as candidate business objects under the target business. Among them, the hot business objects can be understood as business objects that are more concerned or popular among the general public, and can be determined by the number or frequency of user visits to the business objects, and the number or frequency of user clicks. It is understandable that since most users maintain a certain interest and curiosity in hot business objects, the user's acceptance of hot business objects is usually relatively high, that is, hot business objects have a certain probability of recommendation. Therefore, hot business objects can be used as candidate business objects for subsequent further evaluation, which improves the accuracy of business object recommendations while also improving the recommendation efficiency.

[0054] As another way, the electronic device can preliminarily determine the recommended business object based on the attention behavior data generated by the user under the target business as a candidate business object under the target business. It can be understood that since the preliminarily determined recommended business object is already an object to be recommended determined based on the attention behavior data under the target business, the user will have a certain degree of acceptance of the preliminarily determined business object, that is, the preliminarily determined business object has a certain probability of recommendation, so the business object can be used as a candidate business object for subsequent further evaluation to screen out business objects with higher acceptance for recommendation, thereby improving the accuracy of business object recommendation and the efficiency of recommendation.

[0055] In some embodiments, the user's attention behavior under the target business can be labeled, so that the user's interest preferences under the target business type can be classified. As a way, the target business can pre-set corresponding business tags for each business object, and the business tag can reflect the style type characteristics of the business object, so that when the user pays attention to the business object, the attention behavior can be marked with the business tag corresponding to the business object. Since the business tags pre-set by the target business may be incomplete or incorrect, as another way, the business tags can also be set and modified by the user himself. The specific tag settings are not limited here.

[0056] In some embodiments, the business tags set for different business types may be the same, different, or partially the same, which is not limited here. In an embodiment of the present application, when the electronic device obtains at least one candidate business object under the target business, it can also obtain at least one target tag corresponding to each candidate business object to determine the style type characteristics of the candidate business object. Among them, the target tag can be the business tag set corresponding to the target business, which can be determined according to the style type characteristics of the candidate business object.

[0057] It can be understood that when a candidate business object is a business object in the target business that has a certain recommendation probability for a business user, at least one target tag corresponding to the candidate business object can be used to reflect the business user's certain interest preferences in the target business, but it cannot fully reflect the user's actual interest preferences, so further recommendation evaluation of the candidate business object is required later.

[0058] In some embodiments, a tag list corresponding to the attention behavior can also be established in the above-mentioned general attention system, so that the user's interest preference can be determined based on the tag list. As one way, the tag list can be stored corresponding to the attention data. Exemplarily, when the attention relationship data of a user of a business is stored, the storage correspondence can be [business identification ID] -> [follower identification ID] -> [attended object identification ID] -> [tag 1, tag 2, tag 3, ..., tag n].

[0059] Step S120: acquiring at least one reference tag generated by a service user of the target service in a reference service, wherein the reference service is different from the target service, and the reference tag is determined according to the service user's attention behavior in the reference service.

[0060] In an embodiment of the present application, the electronic device may obtain at least one reference tag generated by a service user of the target service in a reference service to determine the interest preference reflected by the user in the reference service. The reference service may be a service used to provide a recommendation basis for a candidate service object of the target service, and the reference service is different from the target service. In some embodiments, the reference service may be one or more, which is not limited here.

[0061] It is understandable that since the universal attention system uniformly stores and processes attention data of different services, it is equivalent to connecting attention behavior data between different services. Therefore, the electronic device can obtain the attention behavior data of the user in different services based on the universal attention system, and can determine the service tags generated by the user in different services based on the attention behavior data, and then determine the user's interest preferences in different services based on the service tags.

[0062] Among them, the business user of the target business can be understood as the user who uses the target business, and the reference tag can be the business tag set corresponding to the reference business, which can be determined according to the style type characteristics of the business object that the business user pays attention to in the reference business. As a way, the electronic device can obtain the attention behavior data generated by the business user of the target business in the reference business, and then determine the corresponding business tag list based on the attention behavior data, and at least one reference tag generated by the business user in the reference business can be obtained based on the business tag list. In other words, the business user of the target business not only uses the target business, but also uses the reference business.

[0063] In some embodiments, the reference service may be a service category that is the same or similar to the target service. For example, when the target service is a paid reading service for electronic books, the reference service may be a book attention service for novels, which is also a text service category. In other embodiments, the reference service may be a service category that is different or not similar to the target service. For example, when the target service is a video service, the reference service may be a book attention service for novels, one is a video service category, and the other is a text service category. The specific relationship between the reference service and the target service is not limited here.

[0064] Step S130: determining a recommendation score for each of the candidate business objects according to the at least one reference tag and the at least one target tag.

[0065] In the embodiment of the present application, the electronic device can determine the recommendation score of each candidate business object in the at least one candidate business object according to the at least one reference tag and the at least one target tag obtained above. In this way, the user's acceptance of each candidate business object under the target business can be comprehensively evaluated according to the user's interest preferences under other business types, and accurate recommendations can be made according to the evaluation results.

[0066] It can be understood that since the above-mentioned at least one reference tag can be used to characterize the interest preferences reflected by the user in the reference business, and the above-mentioned at least one target tag can be used to characterize the interest preferences reflected by the user in the target business, therefore, the interests and hobbies reflected by the user in different businesses of the reference business and the target business can be comprehensively considered to evaluate the recommendation accuracy of each candidate business object, and obtain the recommendation score of each candidate business object, so as to determine whether to recommend the candidate business object based on the recommendation score.

[0067] In some embodiments, for each candidate business object, when the interests and hobbies reflected by the user in the reference business and the target business are similar, it is considered that the recommendation accuracy of the candidate business object is relatively high, and the probability that the user accepts the candidate business object is relatively high, so the recommendation score of the candidate business object is determined to be relatively high. Similarly, when the interests and hobbies reflected by the user in the reference business and the target business are very different, it is considered that the recommendation accuracy of the candidate business object is relatively low, and the probability that the user accepts the candidate business object is relatively low, so the recommendation score of the candidate business object is determined to be relatively low.

[0068] In some embodiments, the electronic device may determine at least one target tag corresponding to each candidate business object, and then determine the recommendation score of the candidate business object based on the matching degree of at least one reference tag and the at least one target tag, thereby obtaining the recommendation score of each candidate business object. The matching degree of at least one reference tag and the at least one target tag may be the matching degree of importance of the tags or the matching degree of similarity of the tags, which is not limited here. For example, it may also be the comprehensive matching degree of the importance and similarity of the tags.

[0069] In some embodiments, the higher the recommendation score of a candidate business object, the greater the probability that the candidate business object is accepted by the business user, or the lower the recommendation score of a candidate business object, the greater the probability that the candidate business object is accepted by the business user. This is not limited here and can be reasonably set according to actual needs.

[0070] Step S140: determining a target business object from the at least one candidate business object according to the recommendation score and recommending the target business object to the business user.

[0071] In the embodiment of the present application, when the recommendation score of each candidate business object is obtained, the target business object recommended to the business user can be determined from at least one candidate business object according to the recommendation score, and the target business object can be recommended. It can be understood that when the recommendation score of the candidate business object indicates that the probability of the candidate business object being accepted by the business user is relatively high, the candidate business object can be recommended, and when the recommendation score of the candidate business object indicates that the probability of the candidate business object being accepted by the business user is relatively low, the candidate business object may not be recommended.

[0072] In some embodiments, at least one candidate business object may be sorted according to the recommendation score of each candidate business object to obtain a sorting result for each candidate business object, and then a target business object may be selected from at least one candidate business object according to the sorting result and recommended to the business user. Optionally, the target business object pushed to the business user each time may be different to reduce user disgust. Sorting at least one candidate business object may be sorting at least one candidate business object in the order of recommendation score from small to large or from large to small, which is not limited here.

[0073] In some embodiments, the candidate business object that is ranked higher or lower in the ranking result may be determined as the target business object recommended to the business user. Specifically, when the recommendation score of the candidate business object is higher, the probability that the candidate business object is accepted by the business user is greater, and the candidate business object that is ranked higher in the ranking result may be determined as the target business object recommended to the business user; when the recommendation score of the candidate business object is lower, the probability that the candidate business object is accepted by the business user is greater, and the candidate business object that is ranked lower in the ranking result may be determined as the target business object recommended to the business user. This is not limited here, and can be reasonably set according to actual needs.

[0074] In other embodiments, the candidate business object with the highest or lowest recommendation score may be determined as the target business object recommended to the business user. Specifically, when the recommendation score of the candidate business object is higher, the probability that the candidate business object is accepted by the business user is greater, and the candidate business object with the highest recommendation score may be determined as the target business object recommended to the business user; when the recommendation score of the candidate business object is lower, the probability that the candidate business object is accepted by the business user is greater, and the candidate business object with the lowest recommendation score may be determined as the target business object recommended to the business user. This is not limited here, and can be reasonably set according to actual needs.

[0075] In some embodiments, multiple businesses can also apply to establish an association relationship, so that when recommending content to business users under the target business, the associated business of the target business can be directly used as the above-mentioned reference business based on the association relationship, so as to achieve content recommendation with reference to the attention data of the associated business. Specifically, when a business applies to establish an association relationship, it can report the business object list and business object type of each business. The business system can establish the user's hobby tag data based on the user's attention behavior data in the business. When the business party requests to recommend data for the user, the business system can calculate based on the user's hobby tag data and recommend related types of items. Therefore, when more associated businesses are connected to the system, more and more comprehensive hobby tags can be added to users, and recommended items can also be expanded to a larger range.

[0076] For example, a novel website (marker 1001), a video website (marker 1002), and a game platform (marker 1003) join the general attention system as business parties. User A follows a number of novels and authors on the novel website. The video website business requests recommended data, and the system can recommend videos (interview records, TV series, movies, variety shows) adapted from the novels of the followed writers, or videos of the same type. The game platform business requests recommended data, and the system can recommend games (web games, mobile games) adapted from the novels of the followed writers, and other content.

[0077] The business object recommendation method provided in the embodiment of the present application obtains at least one candidate business object under the target business, wherein each candidate business object corresponds to at least one target tag, and obtains at least one reference tag generated by the business user of the target business in the reference business, wherein the reference business is different from the target business, and the reference tag is determined according to the attention behavior of the business user in the reference business, and then the recommendation score of each candidate business object is determined according to the at least one reference tag and the at least one target tag, so as to determine and recommend the target business object recommended to the business user from the at least one candidate business object according to the recommendation score. The present application can classify the interests and hobbies of the user under different business types by labeling the attention behavior of the user under different business types, so that when recommending the business object under the target business to the business user under the target business, it can comprehensively evaluate the acceptance of the business user to each candidate business object according to the interests and hobbies of the business user under other business types, so that according to the evaluation results, accurate recommendation of business objects for the business user can be achieved.

[0078] See also Figure 6 , Figure 6 A flow chart of a business object recommendation method provided by another embodiment of the present application is shown, which can be applied to electronic devices. The business object recommendation method may include:

[0079] Step S210: obtaining at least one candidate business object under the target business, each of the candidate business objects corresponding to at least one target tag.

[0080] Step S220: Acquire at least one reference tag generated by a service user of the target service in a reference service.

[0081] In the embodiment of the present application, step S210 and step S210 can refer to the aforementioned embodiment and will not be repeated here.

[0082] Step S230: for each of the target tags, determine a designated reference tag matching the target tag from the at least one reference tag.

[0083] In some embodiments, after obtaining the above-mentioned at least one target tag and at least one reference tag, the electronic device can determine, for each of the target tags, a designated reference tag that matches the target tag from at least one reference tag to determine whether the target tag of the target business appears in the reference tags generated by the user in the reference business.

[0084] It is understandable that, since different businesses may have different business tags, the target tag of the target business and the reference tag of the reference business may have different business tags. When there is a designated reference tag matching the target tag from at least one reference tag, it can be considered that the user has the same or similar interests and hobbies in different businesses of the reference business and the target business. Therefore, the recommendation score of each candidate business object can be determined according to the importance of the same or similar interests and hobbies in different businesses. Among them, the designated reference tag matching the target tag can be a tag that is the same or similar to the target tag.

[0085] Specifically, the electronic device can determine at least one target tag corresponding to each candidate business object, and determine a designated reference tag matching the target tag from at least one reference tag for each target tag in the at least one target tag, thereby obtaining a designated reference tag corresponding to each candidate business object. The designated reference tag corresponding to each candidate business object can be one, indicating that the user has a same or similar interest in different businesses of the reference business and the target business; or multiple, indicating that the user has multiple same or similar interests in different businesses of the reference business and the target business, which is not limited here.

[0086] Step S240: Obtain the importance score of the designated reference tag in the reference service as the first importance score of the target tag in the reference service.

[0087] In some embodiments, after obtaining a designated reference tag that matches the target tag, the importance score of the designated reference tag in the reference business can be obtained as the first importance score of the target tag in the reference business, thereby obtaining the importance of the interests shared by the reference business and the target business in the reference business.

[0088] Among them, in a business, the importance score of each business tag generated by a user's attention behavior can be obtained by first obtaining a list of all tags corresponding to all the attention behaviors generated by the user in the business, and then obtaining the number of tags in all tag lists as the total number of reference tags of the user in the business. Then, the number of occurrences of each business tag, that is, the total number of each business tag, can be determined based on all tag lists. By obtaining the ratio of the total number of each business tag to the total number of reference tags of the user in the business, the proportion of each business tag in the total number of reference tags in the business can be obtained, and then the ratio of each business tag can be used as the importance score of each business tag in the business. That is, the importance score of a certain business tag = the number of occurrences of a certain business tag / the number of occurrences of all tags.

[0089] For example, a user generates two business tags in the novel attention business: urban romance and fairy tales. In this business, the total number of times the user tags business objects in the total attention behavior is 10,000, of which the number of times the "urban romance" tag is 5,000 and the number of times the "fairy tales" tag is 1,000. The importance score of the "urban romance" tag is 0.50 (5,000 / 10,000), and the importance score of the "fairy tales" tag is 0.10 (1,000 / 10,000).

[0090] Specifically, all tag lists corresponding to all attention behaviors generated by the business user in the reference business can be obtained, and then the number of tags in all tag lists can be obtained as the total number of reference tags of the business user in the reference business. The electronic device can then determine the number of occurrences of the specified reference tag based on all tag lists, that is, the total number of specified reference tags. The ratio of the total number of specified reference tags to the total number of reference tags of the business user in the reference business is obtained as the importance score of the specified reference tag in the reference business.

[0091] In some embodiments, for each target tag, when there is no designated reference tag matching the target tag in at least one reference tag, it can be considered that the target tag of the target business does not appear in the reference tags generated by the user in the reference business. At this time, a preset default importance score can also be obtained as the first importance score of the target tag in the reference business. Among them, the default importance score can be reasonably set according to the specific situation, for example, it can be set to 0.01. In this way, the importance score of each target tag of the candidate business object can be evaluated, avoiding the accuracy of the subsequent recommendation evaluation affected by data missing, and improving the accuracy of business object recommendation.

[0092] Step S250: Determine a second importance score of the target tag in the target business.

[0093] In some embodiments, the electronic device may use the second importance score of the target tag in the target business to obtain the importance of the interests and hobbies shared by the reference business and the target business in the target business, so as to determine whether the recommendation of the candidate business object is accurate based on the importance of the interests and hobbies in different businesses.

[0094] Specifically, it can be to obtain all the tag lists corresponding to all the attention behaviors generated by the business user in the target business, and then obtain the number of tags in all tag lists as the total number of reference tags of the business user in the target business. Then the electronic device can determine the number of occurrences of the target tag based on all tag lists, that is, the total number of target tags. The ratio of the total number of target tags to the total number of reference tags of the business user in the target business is obtained as the second importance score of the target tag in the target business.

[0095] Step S260: Determine a recommendation score for each of the candidate business objects according to the first importance score and the second importance score.

[0096] In some embodiments, the recommendation score of each candidate business object may be determined based on the difference between the first importance score and the second importance score. Specifically, the electronic device may determine, for each candidate business object, at least one target tag corresponding to the candidate business object, and for each target tag in the at least one target tag, determine the difference between the first importance score of each target tag in the reference business and the second importance score of each target tag in the target business, so as to obtain the difference in importance of different interest preferences of the candidate business object in different businesses, thereby determining the recommendation score of each candidate business object.

[0097] As one way, the recommendation score of each candidate business object may be determined based on the difference between the first importance score and the second importance score. In order to make the difference more obvious, as another way, the recommendation score of each candidate business object may be determined based on the power of the difference between the first importance score and the second importance score. The power may be 2, i.e., a square.

[0098] In some embodiments, when the difference between the first importance score of each target tag in the reference business and the second importance score of the target business is relatively large, the recommendation score of the candidate business object can be relatively large, and when the difference between the first importance score of each target tag in the reference business and the second importance score of the target business is relatively small, the recommendation score of the candidate business object can be relatively small. This is not limited here, and can be reasonably set according to the specific situation.

[0099] It can be understood that when the difference between the importance scores of the target tag in the reference business and in the target business is relatively large, it can be considered that the importance of the target tag in different businesses is different, that is, the style type reflected by the target tag has different degrees of user interest and preference in different businesses, and the probability of the candidate business object corresponding to the target tag being accepted by the user is uncertain. Therefore, the recommendation priority of the candidate business object can be put back to correspond to a lower recommendation priority. In other words, the larger the recommendation score of the candidate business object, the lower the recommendation priority.

[0100] On the contrary, when the difference between the first importance score of each target tag in the reference business and the second importance score in the target business is relatively small, it can be considered that the importance of the target tag in different businesses is the same or similar, that is, the style type reflected by the target tag has the same degree of user interest and preference in different businesses. It can be considered that the candidate business object determined by the target business has basically reflected the actual interest preference of the user, and the candidate business object corresponding to the target tag is accurately recommended. Therefore, the recommendation priority of the candidate business object can be put forward to correspond to a higher recommendation priority. In other words, the smaller the recommendation score of the candidate business object is, the higher the recommendation priority is.

[0101] For example, suppose reference service A is a novel book following service, and user User1 follows book 1, whose reference tags are "campus romance" and "danmei". Target service B is a paid e-book reading service, and user User1 comes to service B. The candidate e-books recommended by service B to the user are:

[0102] Book recom_book_aa, its corresponding target tags are "campus romance" and "danmei";

[0103] The book recom_book_bb has the corresponding target tag "Eastern Fantasy".

[0104] Assume that in reference business A, the importance score of the "Campus Romance" tag is 0.3, the importance score of the "Danmei" tag is 0.1, and the user User1 with the "Oriental Fantasy" tag is not generated (does not exist) in business A, and uses the default importance score of 0.01; while in target business B, the importance score of the "Campus Romance" tag is 0.25, the importance score of the "Danmei" tag is 0.15, and the importance score of the "Oriental Fantasy" tag is 0.1.

[0105] The electronic device can obtain the recommendation score of the candidate book recom_book_aa as follows: (0.3-0.25) 2 +(0.1-0.15) 2 =0.005, the recommendation score of the candidate book recom_book_bb is: (0.1-0.01) 2 =0.0081. It can be seen that the recommendation score of the candidate book recom_book_aa is relatively small. It can be considered that the importance of the business tags of the candidate book recom_book_aa in different businesses is relatively small, which has basically reflected the actual interest preference of the user, and it can be considered that the probability of being accepted by the user is higher. The recommendation score of the candidate book recom_book_bb is relatively large. It can be considered that the importance of the business tags of the candidate book recom_book_bb in different businesses is relatively large. The book may not reflect the actual interest preference of the user, and it can be considered that the probability of being accepted by the user is lower. Therefore, the candidate book recom_book_aa can be preferentially selected to recommend to user User1 in the future.

[0106] Step S270: Determine a target business object from the at least one candidate business object according to the recommendation score and recommend it to the business user.

[0107] In the embodiment of the present application, step S270 can refer to the aforementioned embodiment and will not be described again here.

[0108] The business object recommendation method provided by the embodiment of the present application obtains at least one candidate business object under the target business, wherein each candidate business object corresponds to at least one target tag, and obtains at least one reference tag generated by the business user of the target business in the reference business, wherein the reference business is different from the target business, and the reference tag is determined according to the attention behavior of the business user in the reference business, and then for each target tag, a specified reference tag matching the target tag is determined from the above-mentioned at least one reference tag to obtain the importance score of the specified reference tag in the reference business as the first importance score of the target tag in the reference business, and the second importance score of the target tag in the target business is determined, so as to determine the recommendation score of each candidate business object based on the first importance score and the second importance score, so as to determine and recommend the target business object recommended to the business user from the above-mentioned at least one candidate business object based on the recommendation score. The present application labels the user's attention behaviors under different business types and generates importance scores for each label. It can classify the user's interests and hobbies under different business types and the importance of each interest and hobby. Therefore, when recommending business objects under the target business to business users under the target business, it can comprehensively evaluate the business user's acceptance of each candidate business object based on the importance of the business user's interests and hobbies under other business types, and thus, based on the evaluation results, achieve accurate recommendation of business objects for the business user.

[0109] See also Figure 7 , Figure 7 A flow chart of a business object recommendation method provided by another embodiment of the present application is shown, which can be applied to electronic devices. The business object recommendation method may include:

[0110] Step S310: obtaining at least one candidate business object under the target business, each of the candidate business objects corresponding to at least one target tag.

[0111] Step S320: Acquire at least one reference tag generated by a service user of the target service in a reference service.

[0112] Step S330: Determine the number of target tags present in the at least one reference tag from the at least one target tag.

[0113] In some embodiments, after obtaining the above-mentioned at least one target tag and at least one reference tag, the electronic device can also determine the recommendation score of each candidate business object according to the tag similarity between the at least one target tag and the at least one reference tag. Specifically, the number of target tags existing in at least one reference tag can be determined from at least one target tag to determine the number of the same or similar interest tags of the user in different businesses of the reference business and the target business, so that the recommendation score of each candidate business object can be determined according to the number of the same or similar interests.

[0114] Specifically, the electronic device can determine at least one target tag corresponding to each candidate business object, and determine the number of target tags existing in at least one reference tag for the at least one target tag, thereby obtaining the number of tags corresponding to each candidate business object. It can be determined whether each target tag in the at least one target tag exists in at least one reference tag, and if so, the number of tags is increased by 1.

[0115] Step S340: Determine a recommendation score for each of the candidate business objects according to the number of tags.

[0116] In some embodiments, the recommendation score of each candidate business object may be determined based on the number of tags corresponding to each candidate business object. As one approach, the larger the number of tags corresponding to a candidate business object, the larger the recommendation score of the candidate business object may be, and the smaller the number of tags corresponding to a candidate business object, the smaller the recommendation score of the candidate business object may be.

[0117] It can be understood that when there are more target tags in at least one reference tag from at least one target tag, it indicates that the target tag of the candidate business object is more likely to appear in the reference tag generated by the user in the reference business, that is, the target tag of the candidate business object is more matched with the reference tag generated by the user in the reference business, it can be considered that the candidate business object determined by the target business has basically reflected the actual interest preference of the user, and the candidate business object corresponding to the target tag is accurately recommended, so the recommendation priority of the candidate business object can be put forward to correspond to a higher recommendation priority. In other words, when the recommendation score of the candidate business object is relatively large, the recommendation priority is higher.

[0118] On the contrary, when the number of target tags that exist in at least one reference tag from at least one target tag is smaller, it indicates that the target tag of the candidate business object has a lower probability of appearing in the reference tag generated by the user in the reference business, that is, the target tag of the candidate business object is less matched with the reference tag generated by the user in the reference business. It can be considered that the candidate business object determined by the target business does not fully reflect the actual interest preference of the user, and the recommendation of the candidate business object corresponding to the target tag is not accurate. Therefore, the recommendation priority of the candidate business object can be put back to correspond to a lower recommendation priority. In other words, the smaller the recommendation score of the candidate business object is, the lower the recommendation priority is.

[0119] For example, in the above example, the candidate data recom_book_aa has a relatively large number of matching tags (2), and a high score, assuming 10; while recom_book_bb has a relatively small number of matching tags (0), and a low score, assuming 5; the one with a higher score is considered to have a greater probability of being accepted by the user. Therefore, the candidate book recom_book_aa can be preferentially selected to be recommended to user User1 in the future.

[0120] Step S350: determining a target business object from the at least one candidate business object according to the recommendation score and recommending the target business object to the business user.

[0121] In the embodiment of the present application, step S350 can refer to the aforementioned embodiment and will not be described again here.

[0122] The business object recommendation method provided by the embodiment of the present application obtains at least one candidate business object under the target business, wherein each candidate business object corresponds to at least one target tag, and obtains at least one reference tag generated by the business user of the target business in the reference business, wherein the reference business is different from the target business, and the reference tag is determined according to the attention behavior of the business user in the reference business, and then from at least one target tag, the number of target tags existing in the above-mentioned at least one reference tag is determined, and the recommendation score of each candidate business object is determined according to the number of tags, so as to determine and recommend the target business object recommended to the business user from the above-mentioned at least one candidate business object according to the recommendation score. The present application can classify the interests and hobbies of the user under different business types by labeling the attention behavior of the user under different business types, so that when recommending the business object under the target business to the business user under the target business, it can comprehensively evaluate the acceptance of the business user to each candidate business object according to the similar interests and hobbies of the business user under other business types, so that according to the evaluation result, the business object for the business user can be accurately recommended.

[0123] See also Figure 8 , Figure 8 A flow chart of a business object recommendation method provided in another embodiment of the present application is shown, which can be applied to electronic devices. The business object recommendation method may include:

[0124] Step S410: obtaining at least one candidate business object under the target business, each of the candidate business objects corresponding to at least one target tag.

[0125] Step S420: Acquire at least one reference tag generated by a service user of the target service in a reference service.

[0126] Step S430: Obtain the business similarity between the target business and the reference business.

[0127] It is understandable that when the business relevance between the reference business and the target business is greater, it can be considered that the possibility that users of the two businesses are interested in and prefer similar items is greater. Therefore, in some embodiments, the recommendation score of each candidate business object can also be comprehensively evaluated by combining the business relevance between the reference business and the target business.

[0128] Since different services have different user groups (user portraits), when the difference between the user groups of two services is small, that is, they are relatively similar, it can be considered that the two services are relatively similar, that is, the correlation between the two services is relatively large. Fig. 9 , step S430 may include:

[0129] Step S431: Determine the target user attributes under the target service and the reference user attributes under the reference service.

[0130] Among them, user attributes are information describing the corresponding user, which may include static attributes such as user age, user gender, user geographic location, user education, marital status, whether or not a paying user, terminal device used, etc., and dynamic attributes such as user consumption level, access time, access duration, etc. The specific user attribute types are not limited here.

[0131] In some embodiments, the user attributes that may be set corresponding to different services may be the same, different, or partially the same, which is not limited here. In an embodiment of the present application, the electronic device can respectively determine the target user attributes of the service user under the target service and the reference user attributes of the service user under the reference service. Among them, the target user attributes can be the user attributes set corresponding to the target service, and the reference user attributes can be the user attributes set corresponding to the reference service.

[0132] Step S432: Determine the service similarity between the target service and the reference service according to the matching degree between the target user attribute and the reference user attribute.

[0133] In some embodiments, after acquiring the target user attributes under the target service and the reference user attributes under the reference service, the electronic device may determine the service similarity between the target service and the reference service according to the matching degree between the target user attributes and the reference user attributes.

[0134] In some embodiments, the business similarity between the target business and the reference business may be determined based on the distribution of each user attribute in the target user attributes and the reference user attributes. Since the target user attributes may be different from the reference user attributes, some target user attributes under the target business may not exist in the reference business, and the distribution of non-existent user attributes cannot be determined. Therefore, as an implementation method, the business similarity between the target business and the reference business may be determined based on the distribution of common attributes in the target user attributes and the reference user attributes. For details, please refer to Fig.10 , step S432 may include:

[0135] Step S4321: Acquire the common user attributes of the target user attributes and the reference user attributes.

[0136] As one way, the electronic device can determine whether each target user attribute of the target service exists in the reference user attribute of the reference service. If so, the target user attribute can be considered as a common user attribute between the target user attribute and the reference user attribute, thereby obtaining all common user attributes between the target user attribute and the reference user attribute. The common user attribute can be one or more. As another way, the electronic device can also determine whether each reference user attribute of the reference service exists in the target user attribute of the target service. If so, the reference user attribute can be considered as a common user attribute between the target user attribute and the reference user attribute.

[0137] Step S4322: Determine a third importance score of the common user attribute in the target service and a fourth importance score of the common user attribute in the reference service.

[0138] In some embodiments, after obtaining the common user attributes of the target user attributes and the reference user attributes, the third importance score of the common user attributes in the target business and the fourth importance score of the common user attributes in the reference business can be determined, thereby obtaining the importance of the common user attributes of the reference business and the target business in different businesses.

[0139] In a business, the importance score of a user attribute can be obtained by first obtaining the total number of business users in the business, then obtaining all business users with the user attribute, and obtaining the ratio of all business users with the user attribute to the total number of business users to obtain the proportion of the user attribute in the business, and then the ratio of each user attribute can be used as the importance score of each user attribute in the business. That is, the importance score of a user attribute = the number of business users of the user attribute / the total number of business users.

[0140] Specifically, for each common user attribute, the ratio of all service users with the common user attribute in the target service to the total number of service users in the target service may be obtained as the third importance score of the common user attribute in the target service, and the ratio of all service users with the common user attribute in the reference service to the total number of service users in the reference service may be obtained as the fourth importance score of the common user attribute in the reference service. Thus, the importance score of each common user attribute in the target service and the reference service may be obtained.

[0141] Step S4323: Determine the business similarity between the target business and the reference business according to the third importance score and the fourth importance score.

[0142] The business similarity between the target business and the reference business can be obtained according to the cosine similarity algorithm, which can use the cosine value of the angle between two vectors in the vector space as a measure of the difference between two individuals. The cosine similarity algorithm is as follows:

[0143]

[0144] Among them, the business users have n common user attributes in the target business X and the reference business Y, i is the i-th common user attribute, X i is the third importance score of the i-th common user attribute in the target business X, Y i is the fourth importance score of the i-th common user attribute in the reference business Y. For example, when the common user attribute is a bachelor's degree, the number of business users with this bachelor's degree user attribute in the target business X is 1500 (the total number of business users is 10000), and the number of business users with this bachelor's degree user attribute in business Y is 2000 (the total number of business users is 20000), then X(i) = 0.15, Y(i) = 0.10.

[0145] According to the above formula, the business correlation cosθ can be calculated, and its value range is 0 to 1. It can be understood that the larger the value is and the closer it is to 1, the closer the angle is to 0 degrees, and the more similar the target business X and the reference business Y are, the greater the value of their reference.

[0146] Step S440: determining a recommendation score for each of the candidate business objects according to the matching degree between the at least one reference tag and the at least one target tag, and the business similarity.

[0147] In some embodiments, after determining the business similarity between the target business and the reference business, the recommendation score of each candidate business object can be determined according to the matching degree between at least one reference tag and at least one target tag, as well as the business similarity. The higher the business similarity and the higher the matching degree, the higher the recommendation score of the candidate business object corresponding to the at least one target tag.

[0148] In some embodiments, the degree of match between the at least one reference tag and the at least one target tag may be the degree of match between the first importance score and the second importance score of the aforementioned embodiment. As a method, the business similarity may be compared with the difference between the first importance score and the second importance score to obtain the recommendation score of each candidate business object. The specific formula for calculating the recommendation score may be:

[0149] score=r / ∑(W(Xi)-W(Yi)) 2 , i=1...n

[0150] Among them, there are n target tags corresponding to the candidate business objects, the business relevance between the target business X and the reference business Y is r, the i value is the i-th target tag of the candidate business object, W(Xi) is the second importance score of the i-th target tag in the target business X, and W(Yi) is the first importance score of the i-th target tag in the reference business Y.

[0151] It is understandable that when the business similarity is higher and the difference between the first importance score and the second importance score is smaller, that is, the larger the numerator and the smaller the denominator, the greater the overall recommendation score. In other words, the more similar the target business X and the reference business Y are, the greater the possibility that users of the two businesses are interested in and prefer similar items, the higher the reference value of the reference business, and at the same time, when the importance of the target tags corresponding to the candidate business objects in different businesses is the same or similar, it can be considered that it has basically reflected the actual interests and preferences of the user, the greater the probability of being accepted by the user, and the higher the recommendation accuracy of the candidate business object, and therefore, the higher the recommendation score of the candidate business object.

[0152] For example, in the above example, assuming that the correlation between the reference business A and the target business B is 0.8, the electronic device can obtain the recommendation score of the candidate book recom_book_aa as: 0.8 / ((0.3-0.25) 2 +(0.1-0.15) 2 )=160, the recommendation score of the candidate book recom_book_bb is: 0.8 / ((0.1-0.01) 2 )≈98.8. It can be seen that the recommendation score of the candidate book recom_book_aa is relatively large, and it can be considered that the probability that the candidate book recom_book_aa is accepted by the user is greater. Therefore, the candidate book recom_book_aa can be preferentially selected to be recommended to the user User1 in the future.

[0153] Step S450: determining a target business object from the at least one candidate business object according to the recommendation score and recommending the target business object to the business user.

[0154] In the embodiment of the present application, step S450 can refer to the aforementioned embodiment and will not be described again here.

[0155] The business object recommendation method provided by the embodiment of the present application obtains at least one candidate business object under the target business, wherein each candidate business object corresponds to at least one target tag, and obtains at least one reference tag generated by the business user of the target business in the reference business, wherein the reference business is different from the target business, and the reference tag is determined according to the attention behavior of the business user in the reference business, and then obtains the business similarity between the target business and the reference business, so as to determine the recommendation score of each of the candidate business objects according to the matching degree between at least one reference tag and at least one target tag, and the business similarity, and finally, according to the recommendation score, determines and recommends the target business object recommended to the business user from the above-mentioned at least one candidate business object. Among them, the higher the business similarity and the higher the matching degree between at least one reference tag and at least one target tag, the higher the recommendation score of the candidate business object corresponding to the at least one target tag. This application can classify the user's interests and hobbies in different business types by labeling the user's attention behaviors in different business types. Therefore, when recommending business objects under the target business to business users under the target business, it can comprehensively evaluate the business user's acceptance of each candidate business object based on the business user's similar interests and hobbies in other business types, as well as the business relevance of the target business to other business types, so as to achieve accurate recommendation of business objects for the business user based on the evaluation results.

[0156] See also Fig.11 , Fig.11A flow chart of another service object recommendation method provided in an embodiment of the present application is shown, which can be applied to an electronic device. The service object recommendation method may include:

[0157] Step S510: determining similar service users of each service user according to the user attributes of each service user under the target service.

[0158] Since users with similar user attributes (similar user portraits) are more likely to like similar items, in some embodiments, the electronic device can recommend business objects to users based on the similarity of user attributes. Specifically, the electronic device can determine similar business users of each business user under the target business based on the user attributes of each business user under the target business, and thus recommend business objects of interest and preference to business users under the target business based on the business objects of interest and preference of similar business users.

[0159] In some embodiments, the user attributes of each service user under the target service can be obtained, and the attribute similarity between all service users under the target service can be determined to obtain a similarity matrix, which can be updated regularly. When the target service needs to recommend content to a designated service user under the target service, a list of similar service users of the designated service user can be obtained based on the similarity matrix to determine similar service users of the designated service user based on the similar service user category.

[0160] As a way, the attribute similarity between two users can be obtained based on the cosine similarity algorithm, which can use the cosine value of the angle between two vectors in the vector space as a measure of the size of the difference between two individuals. Specifically, for users a and b, the cosine similarity algorithm is as follows:

[0161]

[0162] Among them, there are n user attributes under the target business; the value i is the i-th user attribute under the target business; X i It is used to indicate whether user a has the i-th user attribute. If yes, then X i The value of X is 1, if it does not exist i The value of Y is 0; i It is used to indicate whether user b has the i-th user attribute. If yes, then Y i The value is 1, if it does not exist, Y i The value of is 0.

[0163] According to the above formula, the attribute similarity cosθ between two users can be calculated, and its value range is 0 to 1. It can be understood that the larger the value is and the closer it is to 1, the closer the angle is to 0 degrees, and the more similar users a and b are, the greater the value of mutual reference.

[0164] In some embodiments, all business users whose attribute similarity values ​​are less than a preset similarity value may be regarded as similar business users of the designated business user, or a list of similar business users may be sorted in ascending order according to the attribute similarity values, and then a preset number of business users with higher similarity at the top of the sort may be obtained as similar business users of the designated business user. The preset similarity value and the preset number may be set according to actual needs and are not limited here. For example, the preset similarity value may be 0.3, and the preset number may be 100.

[0165] In some embodiments, the business object to be recommended may be determined from the business objects followed by similar business users in the target business according to the number of business objects carried in the request parameter. The request parameter may be a request initiated by the target business connected to the universal attention system of the present application, which may include: the recommended designated business user ID, the similarity ranking threshold (i.e., the above-mentioned preset number), and the number of recommended business objects.

[0166] Step S520: determining at least one candidate business object under the target business according to the business object that the similar business user is interested in in the target business, each of the candidate business objects corresponding to at least one target tag.

[0167] In some embodiments, after similar business users of each business user are determined, business objects that similar business users are interested in in the target business may be obtained as at least one candidate business object under the aforementioned target business, so as to further determine the recommendation score in the aforementioned embodiment.

[0168] It is understandable that since the business objects recommended to the specified business user are determined only by the attribute similarity between all business users under the target business, it is very likely that they do not fully reflect the actual interest preferences of the specified business user. Therefore, in the embodiment of the present application, the interest preferences reflected by the specified business user in other businesses can be obtained to comprehensively consider the interest preferences of the user in different businesses to accurately recommend business objects under the target business to the specified business user.

[0169] Step S530: obtaining at least one reference tag generated by a service user of the target service in a reference service, wherein the reference service is different from the target service, and the reference tag is determined according to the service user's attention behavior in the reference service.

[0170] Step S540: determining a recommendation score for each of the candidate business objects according to the at least one reference tag and the at least one target tag.

[0171] Step S550: determining a target business object from the at least one candidate business object according to the recommendation score and recommending the target business object to the business user.

[0172] In the embodiment of the present application, steps S530 to S550 may refer to the aforementioned embodiment and will not be described in detail here.

[0173] The business object recommendation method provided in the embodiment of the present application determines similar business users of each business user through the user attributes of each business user under the target business, so as to determine at least one candidate business object under the target business according to the business objects concerned by similar business users in the target business, wherein each candidate business object corresponds to at least one target tag, and obtains at least one reference tag generated by the business user of the target business in the reference business, wherein the reference business is different from the target business, and the reference tag is determined according to the attention behavior of the business user in the reference business, and then the recommendation score of each candidate business object is determined according to the at least one reference tag and the at least one target tag, so as to determine and recommend the target business object recommended to the business user from the at least one candidate business object according to the recommendation score. The present application can classify the interests and hobbies of the user under different business types by labeling the attention behavior of the user under different business types, so that when recommending candidate business objects to the business user according to the business objects concerned by similar business users of each business user in the target business, the acceptance of the candidate business object by the business user can be further comprehensively evaluated according to the interests and hobbies of the business user under other business types, so that the business object for the business user can be accurately recommended according to the evaluation result.

[0174] See also Fig.12 , Fig.12The structural block diagram of a business object recommendation device 400 provided by an embodiment of the present application is shown, and the business object recommendation device 400 is applied to an electronic device. The business object recommendation device 400 includes: a candidate object acquisition module 410, a reference tag acquisition module 420, a recommendation score evaluation module 430, and a target object recommendation module 440. Among them, the candidate object acquisition module 410 is used to acquire at least one candidate business object under the target business, and each of the candidate business objects corresponds to at least one target tag; the reference tag acquisition module 420 is used to acquire at least one reference tag generated by a business user of the target business in a reference business, wherein the reference business is different from the target business, and the reference tag is determined according to the attention behavior of the business user in the reference business; the recommendation score evaluation module 430 is used to determine the recommendation score of each of the candidate business objects according to the at least one reference tag and the at least one target tag; the target object recommendation module 440 is used to determine the target business object from the at least one candidate business object according to the recommendation score and recommend it to the business user.

[0175] In some embodiments, the recommendation score evaluation module 430 may include: a label matching unit, used to determine, for each of the target labels, a designated reference label that matches the target label from the at least one reference label; a first score acquisition unit, used to obtain the importance score of the designated reference label in the reference business as the first importance score of the target label in the reference business; a second score acquisition unit, used to determine the second importance score of the target label in the target business; and a score determination unit, used to determine the recommendation score of each of the candidate business objects based on the first importance score and the second importance score.

[0176] In some embodiments, the business object recommendation device 400 may also include: a default score acquisition module, which is used to obtain a preset default importance score for each of the target tags when there is no designated reference tag matching the target tag in the at least one reference tag, as the first importance score of the target tag in the reference business.

[0177] In some other embodiments, the recommendation score evaluation module 430 may also be specifically used to: determine the number of target tags present in the at least one reference tag from the at least one target tag; and determine the recommendation score of each of the candidate business objects based on the number of tags.

[0178] In some other embodiments, the recommendation score evaluation module 430 may also include: a business evaluation unit, used to obtain the business similarity between the target business and the reference business; a comprehensive evaluation unit, used to determine the recommendation score of each of the candidate business objects based on the matching degree between the at least one reference tag and the at least one target tag, and the business similarity, wherein the higher the business similarity and the higher the matching degree, the higher the recommendation score of the candidate business object corresponding to the at least one target tag.

[0179] In some embodiments, the above-mentioned business evaluation unit may include: an attribute determination subunit, used to determine the target user attributes under the target business, and the reference user attributes under the reference business; an attribute matching subunit, used to determine the business similarity between the target business and the reference business based on the degree of matching between the target user attributes and the reference user attributes.

[0180] In some embodiments, the above-mentioned attribute matching subunit can be specifically used to: obtain the common user attributes between the target user attributes and the reference user attributes; determine the third importance score of the common user attributes in the target business, and the fourth importance score of the common user attributes in the reference business; determine the business similarity between the target business and the reference business based on the third importance score and the fourth importance score.

[0181] In some embodiments, the candidate object acquisition module 410 can be specifically used to: determine similar business users of each business user under the target business based on the user attributes of the business user; determine at least one candidate business object under the target business based on the business objects that the similar business users are interested in in the target business.

[0182] The business object recommendation device provided in the embodiment of the present application is used to implement the corresponding business object recommendation method in the aforementioned method embodiment, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0183] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0184] In several embodiments provided in the present application, the coupling or direct coupling or communication connection between the modules shown or discussed may be an indirect coupling or communication connection through some interfaces, devices or modules, which may be electrical, mechanical or other forms.

[0185] In addition, each functional module in each embodiment of the present application can be integrated into a processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or software functional modules.

[0186] Please refer to Fig.13 , Fig.13 It shows a structural block diagram of an electronic device provided in an embodiment of the present application. The electronic device 700 may be the above-mentioned terminal device, wherein the terminal device may be a user device capable of running an application such as a smart phone, a tablet computer, a laptop computer, a desktop computer, a wearable terminal device, etc. The electronic device 700 may also be the above-mentioned server. The electronic device 700 in the present application may include one or more of the following components: a processor 710, a memory 720, and one or more applications, wherein one or more applications may be stored in the memory 720 and configured to be executed by one or more processors 710, and one or more programs are configured to execute the method as described in the aforementioned method embodiment.

[0187] The processor 710 may include one or more processing cores. The processor 710 uses various interfaces and lines to connect various parts of the entire electronic device 700, and executes various functions and processes data of the electronic device 700 by running or executing instructions, programs, code sets or instruction sets stored in the memory 720, and calling data stored in the memory 720. Optionally, the processor 710 can be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor 710 can integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 710, but may be implemented separately through a communication chip.

[0188] The memory 720 may include a random access memory (RAM) or a read-only memory (ROM). The memory 720 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 720 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the following various method embodiments, etc. The data storage area may also store data created by the electronic device 700 during use, etc.

[0189] Those skilled in the art will understand that Fig.13 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0190] See also Fig.14 , which shows a structural block diagram of a computer-readable storage medium provided in an embodiment of the present application. The computer-readable storage medium 800 stores program codes, which can be called by a processor to execute the method described in the above method embodiment.

[0191] The computer readable storage medium 800 can be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM, a hard disk, or a ROM. Optionally, the computer readable storage medium 800 includes a non-transitory computer-readable storage medium. The computer readable storage medium 800 has storage space for program code 810 that performs any method step of the above method. These program codes can be read from or written to one or more computer program products. The program code 810 can be compressed, for example, in an appropriate form.

[0192] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0193] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A business object recommendation method, characterized in that: The method comprises: Acquire at least one candidate business object under the target business, each of the candidate business objects corresponding to at least one target tag; Acquire at least one reference tag generated by a service user of the target service in a reference service, wherein the reference service is different from the target service, and the reference tag is determined according to a concern behavior of the service user in the reference service; Determine the recommendation score of each of the candidate business objects according to the at least one reference tag and the at least one target tag, wherein the determining the recommendation score of each of the candidate business objects according to the at least one reference tag and the at least one target tag comprises: for each of the target tags, determining a designated reference tag that matches the target tag from the at least one reference tag; obtaining an importance score of the designated reference tag in the reference business as a first importance score of the target tag in the reference business; determining a second importance score of the target tag in the target business; determining the recommendation score of each of the candidate business objects according to the first importance score and the second importance score; wherein the determining the recommendation score of each of the candidate business objects according to the first importance score and the second importance score comprises: determining the recommendation score of each of the candidate business objects according to the power of the difference between the first importance score and the second importance score; According to the recommendation score, a target business object is determined from the at least one candidate business object and recommended to the business user.

2. The method according to claim 1, characterized in that Before determining the recommendation score of each candidate business object according to the first importance score and the second importance score, the method further includes: For each of the target tags, when there is no designated reference tag matching the target tag in the at least one reference tag, a preset default importance score is obtained as a first importance score of the target tag in the reference service.

3. The method according to claim 1, characterized in that The determining, according to the at least one reference tag and the at least one target tag, a recommendation score for each of the candidate business objects comprises: Determining, from the at least one target tag, a tag number of the target tag present in the at least one reference tag; A recommendation score for each of the candidate business objects is determined according to the number of tags.

4. The method according to claim 1, characterized in that The determining, according to the at least one reference tag and the at least one target tag, a recommendation score for each of the candidate business objects comprises: Acquire the business similarity between the target business and the reference business; The recommendation score of each of the candidate business objects is determined based on the degree of matching between the at least one reference tag and the at least one target tag, as well as the business similarity, wherein the higher the business similarity and the higher the degree of matching, the higher the recommendation score of the candidate business object corresponding to the at least one target tag.

5. The method according to claim 4, characterized in that The obtaining of the business similarity between the target business and the reference business includes: Determining target user attributes under the target service and reference user attributes under the reference service; The service similarity between the target service and the reference service is determined according to the matching degree between the target user attribute and the reference user attribute.

6. The method according to claim 5, characterized in that The determining, according to the matching degree between the target user attribute and the reference user attribute, the service similarity between the target service and the reference service comprises: Acquire common user attributes among the target user attributes and the reference user attributes; Determining a third importance score of the common user attribute in the target service and a fourth importance score of the common user attribute in the reference service; The business similarity between the target business and the reference business is determined according to the third importance score and the fourth importance score.

7. The method according to any one of claims 1 to 6, characterized in that: The obtaining of at least one candidate service object under the target service includes: Determine similar business users for each business user according to the user attributes of each business user under the target business; At least one candidate business object under the target business is determined according to the business object that the similar business user is interested in in the target business.

8. A business object recommendation device, characterized in that: The device comprises: A candidate object acquisition module, used to acquire at least one candidate business object under a target business, each of the candidate business objects corresponding to at least one target tag; A reference tag acquisition module, used to acquire at least one reference tag generated by a service user of the target service in a reference service, wherein the reference service is different from the target service, and the reference tag is determined according to the service user's attention behavior in the reference service; A recommendation score evaluation module, configured to determine a recommendation score for each of the candidate business objects according to the at least one reference tag and the at least one target tag, wherein the determining the recommendation score for each of the candidate business objects according to the at least one reference tag and the at least one target tag comprises: for each of the target tags, determining a designated reference tag that matches the target tag from the at least one reference tag; obtaining an importance score of the designated reference tag in the reference business as a first importance score of the target tag in the reference business; determining a second importance score of the target tag in the target business; determining a recommendation score for each of the candidate business objects according to the first importance score and the second importance score; wherein the determining the recommendation score for each of the candidate business objects according to the first importance score and the second importance score comprises: determining the recommendation score for each of the candidate business objects according to a power of a difference between the first importance score and the second importance score; The target object recommendation module is used to determine a target business object from the at least one candidate business object according to the recommendation score and recommend the target business object to the business user.

9. An electronic device, characterized in that: include: one or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to execute the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, which can be called by a processor to execute the method according to any one of claims 1 to 7.

Images