Information recommendation method and device, electronic equipment and storage medium

Abstract

The invention provides an information recommendation method, belongs to the technical field of artificial intelligence, and particularly relates to the technical field of big data, intelligent recommendation and intelligent e-commerce. According to the specific implementation scheme, according to configuration information of a target entity, entity features for the target entity and mode features of corresponding activity modes are determined, and the configuration information represents at least one activity mode, about an object, of the target entity; a target object set corresponding to the target entity is determined according to the interestingness of the multiple objects about the target entity, the target object set comprises at least one target object, and the interestingness of the target object about the target entity is larger than an interestingness threshold value; and pushing the target entity and the corresponding activity mode to at least one target object in the target object set. The invention further provides an information recommendation device, electronic equipment and a storage medium.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, in particular to the field of big data, intelligent recommendation, and intelligent e-commerce technology, and can be used in scenarios such as product recommendation. More specifically, the present disclosure provides an information recommendation method, device, electronic device, and storage medium. Background Art

[0002] With the development of mobile Internet technology, more and more consumers can use different applications to shop, watch videos, etc. At present, in these applications, products and videos of interest can be recommended to users, but the accuracy of the recommendations is low. Summary of the invention

[0003] The present disclosure provides an information recommendation method, apparatus, device and storage medium.

[0004] According to one aspect of the present disclosure, there is provided an information recommendation method, the method comprising: determining entity features and pattern features of corresponding activity patterns for a target entity based on configuration information of the target entity, the configuration information characterizing at least one activity pattern of the target entity with respect to an object; determining a target object set corresponding to the target entity based on the interest of a plurality of objects with respect to the target entity, the target object set comprising at least one target object, the interest of the target object with respect to the target entity being greater than an interest threshold; and pushing the target entity and the corresponding activity pattern to at least one target object in the target object set.

[0005] According to another aspect of the present disclosure, an information recommendation device is provided, which includes: a first determination module, used to determine entity features for a target entity and pattern features of a corresponding activity pattern based on configuration information of the target entity, the configuration information characterizing at least one activity pattern of the target entity with respect to an object; a second determination module, used to determine a target object set corresponding to the target entity based on the interest of multiple objects with respect to the target entity, the target object set including at least one target object, the interest of the target object with respect to the target entity being greater than an interest threshold; and a push module, used to push the target entity and the corresponding activity pattern to at least one target object in the target object set.

[0006] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method provided according to the present disclosure.

[0007] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided. The computer instructions are used to cause a computer to execute the method provided according to the present disclosure.

[0008] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the method provided according to the present disclosure is implemented.

[0009] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0011] Figure 1 The system architecture of the information recommendation method, apparatus, device, medium and program product according to the embodiments of the present disclosure is schematically shown;

[0012] Figure 2 The flowchart of the information recommendation method according to the embodiment of the present disclosure is schematically shown;

[0013] Figure 3 The information recommendation process according to an embodiment of the present disclosure is schematically shown;

[0014] Figure 4 The process of determining an object set according to an embodiment of the present disclosure is schematically shown;

[0015] Figure 5 The following schematically shows an application diagram of the information recommendation method according to an embodiment of the present disclosure;

[0016] Figure 6 A schematic diagram of a first signboard according to an embodiment of the present disclosure is schematically shown;

[0017] Figure 7 A schematic diagram of a second signboard according to an embodiment of the present disclosure is schematically shown;

[0018] Figure 8 is a block diagram of an information recommendation device according to an embodiment of the present disclosure; and

[0019] Fig. 9 is a block diagram of an electronic device to which an information recommendation method can be applied according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0020] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0021] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0022] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0023] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0024] With the development of mobile Internet technology, more and more consumers can use different applications to shop, watch videos, etc. At present, in these applications, products and videos of interest can be recommended to users, but the accuracy of the recommendations is low.

[0025] For example, in the e-commerce business, many merchants join e-commerce platforms to conduct online sales, and consumers can purchase the merchants' products through e-commerce platforms. However, many small and micro merchants lack experience and resources in e-commerce operations and find it difficult to cope with consumers' diverse product demands, resulting in poor product recommendation results. Therefore, they face fierce industry competition pressure.

[0026] In order to ease the competitive pressure on merchants, e-commerce platforms usually provide project support to merchants on the platform. Currently, the support methods for merchants can include one of the following:

[0027] (1) Commodity cards are commission-free, reducing merchants’ operating costs;

[0028] (2) Free entry into the e-commerce platform to solve the problem of difficulty in opening stores for small and medium-sized and individual businesses and alleviate the financial pressure on businesses;

[0029] (3) Merchant-specific identity identification helps merchants gain better traffic exposure.

[0030] However, simply reducing merchant costs and alleviating financial pressure cannot solve the fundamental problems of merchants; and blindly increasing merchant exposure not only cannot effectively help merchants obtain effective users, but also easily reduces the user experience. Therefore, how to more accurately select target user groups and recommend products of interest to target users has become the key to merchant operation strategies.

[0031] To this end, the disclosed embodiment proposes an information recommendation scheme. For example, according to the configuration information of the target entity, the entity features and the mode features of the corresponding activity mode for the target entity are determined, and the configuration information represents at least one activity mode of the target entity with respect to the object; according to the interest of multiple objects with respect to the target entity, a target object set corresponding to the target entity is determined, and the target object set includes at least one target object, and the interest of the target object with respect to the target entity is greater than the interest threshold; and the target entity and the corresponding activity mode are pushed to at least one target object in the target object set.

[0032] According to an embodiment of the present disclosure, entity features and pattern features for screening target objects are determined through the configuration information of the target entity. Target objects are screened out based on the interest of different objects in the target entity corresponding to the entity features and pattern features. And relevant information of the target entity is accurately recommended to the target object. Not only can objects matching the entity be mined more accurately; the connection between the target entity and the target object can also be quickly established, making the connection between the target object and the target entity closer. Thereby, the recommendation results of the target entity are more in line with the needs of users, which helps to improve the efficiency of operations and user experience.

[0033] It should be noted that the information recommendation method proposed in the present disclosure can be applied to scenarios such as e-commerce live broadcast recommendation, intelligent marketing, and value analysis of homogeneous groups, so as to determine the target users of the business, achieve precision marketing, improve the accuracy of the recommendation strategy, and increase the utilization rate of traffic. The information recommendation method provided in the embodiment of the present disclosure does not limit the application scenario.

[0034] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0035] In the technical solution of the present disclosure, the user's authorization or consent is obtained before obtaining or collecting the user's personal information.

[0036] Figure 1The system architecture of the information recommendation method, apparatus, device, medium and program product according to the embodiments of the present disclosure is schematically shown. It should be noted that: Figure 1 What is shown is merely an example of a system architecture to which the embodiments of the present disclosure can be applied, in order to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios.

[0037] like Figure 1 As shown, the system architecture 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, a server 105, and a database 106. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may also be used to provide a medium for a communication link between the server 105 and the database 106. The network 104 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0038] The user may use at least one of the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0039] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0040] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received configuration information and other data, and feed back the processing results (such as obtaining or generating a target object according to the configuration information of the target entity, and pushing the target entity to the target object) to the terminal device.

[0041] The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server can also be a server of a distributed system, or a server combined with blockchain.

[0042] The system architecture 100 may also include one or more databases 106. In some embodiments, these databases may be used to store data and other information. For example, one or more of the databases 106 may be used to store data such as user consumption behavior data, user portraits, user portrait tags, and the like. The data repository 106 may reside in various locations. For example, the data repository used by the server 105 may be local to the server 105, or may be remote from the server 105 and may communicate with the server 105 via a network-based or dedicated connection. The data repository 106 may be of different types. In some embodiments, the data repository used by the server 105 may be a database, such as a relational database. One or more of these databases may store, update, and retrieve data from the database and the database in response to a command.

[0043] In some embodiments, one or more of the databases 106 may also be used by applications to store application data. The databases used by the applications may be different types of databases, such as a key-value store, an object store, or a conventional store backed by a file system.

[0044] It should be noted that the information recommendation method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the information recommendation device provided in the embodiment of the present disclosure can generally be set in the server 105. The information recommendation method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the information recommendation device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0045] Alternatively, the information recommendation method provided in the embodiment of the present disclosure may also be executed by the first terminal device 101, the second terminal device 102, or the third terminal device 103, or may also be executed by other terminal devices different from the first terminal device 101, the second terminal device 102, or the third terminal device 103. Accordingly, the information recommendation apparatus provided in the embodiment of the present disclosure may also be provided in the first terminal device 101, the second terminal device 102, or the third terminal device 103, or may be provided in other terminal devices different from the first terminal device 101, the second terminal device 102, or the third terminal device 103.

[0046] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0047] It should be noted that the sequence numbers of the operations in the following method are only used as representations of the operations for the purpose of description, and should not be regarded as representing the execution order of the operations. Unless explicitly stated, the method does not need to be executed completely in the order shown.

[0048] The above describes the system architecture to which the information recommendation method provided by the present disclosure can be applied. Figure 2 For example, the information recommendation process of the present disclosure is further explained.

[0049] Figure 2 The flowchart of the information recommendation method according to the embodiment of the present disclosure is schematically shown.

[0050] like Figure 2 As shown, the method 200 may include operations S210 to S230.

[0051] In operation S210, entity features for the target entity and mode features of corresponding activity modes are determined according to configuration information of the target entity, the configuration information representing at least one activity mode of the target entity with respect to the object.

[0052] In the embodiment of the present disclosure, the target entity may be any type of entity to be recommended. For example, the target entity may be a merchant, a commodity, a host, advertising information, etc. The embodiment of the present disclosure does not limit this.

[0053] Configuration information can be used to characterize the recommendation strategy of the target entity. Exemplarily, the configuration information may include the recommendation method and recommendation scope of the target entity. For example, the configuration information may be that the e-commerce platform provides financial support to merchant A (target entity), provides coupons to users who purchase merchant A's products, and provides free shipping to users who purchase merchant A's products. The configuration information may also be the traffic support provided by merchant A (target entity) for brand promotion, and merchant A is recommended on the homepage of some applications through relevant platforms for traffic exposure.

[0054] It should be noted that the configuration information can be information provided by the merchant to the recommendation platform, or information provided by the recommendation platform to the merchant. One configuration information can include one or more activity modes. The present disclosure does not limit the number of activity modes.

[0055] The entity feature may be the description information of the target entity. For example, the products sold by merchant A are stationery products. Then the entity features of merchant A may include: stationery, A (brand name of merchant A).

[0056] The activity mode may be information corresponding to the recommended mode (activity mode) in the configuration information of the target entity. The mode feature may be descriptive information of the activity mode. For example, when the configuration information provides a coupon for a user who purchases a product of merchant A, the mode feature may be a coupon.

[0057] For example, the configuration information for down jacket merchant A may be: the e-commerce platform provides coupons to users who purchase products of merchant A. Then the entity features of merchant A may include: down jacket and A, and the mode features may include: coupons.

[0058] In operation S220, a target object set corresponding to the target entity is determined according to the interest of the plurality of objects with respect to the target entity, the target object set including at least one target object whose interest with respect to the target entity is greater than an interest threshold.

[0059] In the disclosed embodiments, the interest level may be the degree to which the object is interested in the target entity. The interest level may be reflected by entity features and pattern features. For example, user 1 may have a first interest level in down jackets; user 1 may have a second interest level in A (brand name); user 1 may have a third interest level in down jacket products with coupons. The degree of interest is reflected differently for different entities. Exemplarily, in the case where the target entity is a product, the interest level may indicate the user's willingness to purchase the product. In the case where the target entity is an anchor, the interest level may indicate whether the user is willing to follow the anchor.

[0060] The interest threshold can be set according to the pattern characteristics. For example, the e-commerce platform can issue coupons to users whose interest in the products of merchant A is greater than 50%. The e-commerce platform can issue coupons to users whose interest in the products of merchant A is greater than 30%. The present disclosure does not limit the size of the interest threshold.

[0061] The target object set is a set consisting of one or more target objects. All target objects in the target object set have interest in the target entity, and the interest of the target object in the target entity is greater than the interest threshold.

[0062] The target object set can be understood as a set formed by users whose interest in target entities, different activity modes, or target entities with different activity modes is greater than 30%. Exemplarily, users 1-10 are more than 30% interested in purchasing stationery products with coupons. Users 11-20 are more than 30% interested in purchasing stationery products with free shipping. Users 21-30 are more than 30% interested in stationery merchant A. Then users 1-30 together constitute the target object set. For target entities, different activity modes, or target entities with different activity modes, where the interest is greater than 30%, there may be duplicate users, in which case the duplicate users can be merged.

[0063] It should be noted that, in the process of determining the target object screening, the more entity features and pattern features are introduced, the higher the degree of adaptation between the determined target object and the target entity.

[0064] For example, when the configuration information of down jacket merchant A is that the e-commerce platform issues coupons to users whose interest in merchant A's products is greater than 10%, users can be screened based on all entity features: down jacket, A and all pattern features: coupon, interest threshold greater than 10%, to obtain users whose interest in brand A down jackets with coupons is greater than 10%.

[0065] In operation S230 , the target entity and the corresponding activity pattern are pushed to at least one target object in the target object set.

[0066] In the disclosed embodiment, after determining the target object set, the information of the target entity with different activity modes can be pushed to the corresponding target object. That is, the target objects determined by the combination of different entity features and mode features are respectively. Exemplarily, user 1-10 has a greater than 30% interest in purchasing stationery products with coupons. Then the information of stationery merchant A with purchase coupons is pushed to user 1-10. User 11-20 has a greater than 30% interest in purchasing stationery products with free shipping. Then the information of stationery merchant A with free shipping is pushed to user 11-20. User 21-30 has a greater than 30% interest in stationery merchant A. Then the information of stationery merchant A is pushed to user 21-30. That is, users who have a certain interest in different activity modes of merchant A are pushed the information of merchant A and / or activity mode information respectively. The information of merchant A can be the product information of merchant A or the store information of merchant A.

[0067] Through the disclosed embodiments, entity features and pattern features used to filter target objects are determined through the configuration information of the target entity. Target objects are filtered out based on the interest of different objects in the target entity corresponding to the entity features and pattern features. And the relevant information of the target entity is accurately recommended to the target object. Not only can objects matching the entity be more accurately excavated; the connection between the target entity and the target object can also be quickly established, making the connection between the target object and the target entity closer. Thereby, the recommendation results of the target entity are more in line with the needs of users, which helps to improve the efficiency of operations and user experience.

[0068] According to the interest of multiple objects in the target entity, a target object set corresponding to the target entity is determined, including: obtaining at least one target feature, the target feature includes an entity feature and a pattern feature; according to the at least one target feature, a candidate object set associated with the at least one target feature is determined from multiple object sets, the candidate object set includes at least one candidate object; when the interest of the candidate object in the target entity is greater than an interest threshold, the candidate object is determined to be a target object, the target object set is obtained, and the information recommendation result is determined.

[0069] The target feature may be determined based on the target entity and the configuration information of the target entity, may include one or more entity features of the target entity, may include one or more pattern features, or may include at least one entity feature and at least one pattern feature.

[0070] As an example, the target feature may also be determined based on the target entity and configuration information of the target entity.

[0071] The object set can be determined by performing association analysis on different objects and different entities.

[0072] The candidate object set may include one or more candidate objects, each of which has interest for the target entity, and the candidate object set includes candidate objects whose interest is greater than or equal to an interest threshold and whose interest is less than the interest threshold.

[0073] In one example, the configuration information of stationery merchant A is to issue coupons to users whose interest in merchant A's products is greater than 10%. Then, users can be screened based on entity features: stationery, A (brand name), and coupons to obtain users 1-50 who are interested in brand A stationery with coupons. Among users 1-50, users 1-30 have an interest in brand A stationery with coupons greater than or equal to 10%, and users 31-50 have an interest in brand A stationery with coupons less than 10%. Then, based on the interest threshold of 10%, users 1-50 are screened, and finally users 1-30 are determined, and merchant A's products and coupons are pushed to users 1-30.

[0074] It can be understood that the candidate objects are obtained by screening the objects through the entity features and pattern features corresponding to the target entity, and then the candidate objects are further screened based on the interest level to obtain the target object. Therefore, setting the target features and interest level thresholds to continuously screen the objects not only improves the accuracy of determining the target object, but also makes the positioning of the target entity activity marketing more accurate and improves the operational efficiency. It can also meet the different needs of users for the same entity and improve the user experience.

[0075] In the embodiments of the present disclosure, information recommendation can be determined based on the target characteristics and interest of the target entity. Figure 3 Take as an example to illustrate the process of determining information recommendation.

[0076] Figure 3 The information recommendation process according to an embodiment of the present disclosure is schematically illustrated.

[0077] like Figure 3 As shown, the configuration information of the target entity is obtained 301, and the entity feature 302 for the target entity and the pattern feature 303 of the corresponding activity mode are determined according to the configuration information of the target entity. At least one target feature is obtained, and the screening condition of the candidate object is constructed. The target feature includes the entity feature 302 and the pattern feature 303. According to the at least one target feature, object matching is performed from a plurality of pre-determined object sets 304, and a candidate object set 305 associated with the at least one target feature is screened out, and the candidate object set 305 includes at least one candidate object. When the interest of the candidate object for the target entity is greater than the interest threshold, the candidate object is determined to be the target object, and the target object set 306 is obtained, and the information recommendation result 307 is determined.

[0078] According to an embodiment of the present disclosure, multiple object sets are determined by the following operations: obtaining first data of multiple objects and second data of multiple entities, the first data representing the descriptive information and behavioral information of the objects, and the second data representing the descriptive information of the entities; generating at least one first label for each object based on the first data, the first label including at least one of the descriptive features and preference features of the object; generating at least one second label for each entity based on the second data, the second label including the entity features and pattern features of the entity; determining multiple object sets based on the first labels and the second labels, the object sets representing the interest of the objects with the first labels in the entities with the second labels.

[0079] The first data may be the description information and behavior information of the object. The description information may be the basic information of the object. For example, the user's age, region, gender, occupation, life stage, income level, etc. The behavior information may be the access information of the object. For example, the content of the user's web browsing, product purchase information, video viewing information, etc. Based on the description information, the description features of the object may be generated. Based on the behavior information, the preference features of the object may be generated.

[0080] Users are aware of and agree to the acquisition and use of First Data, which complies with relevant laws and regulations and does not violate public order and good morals.

[0081] The first tag can be used to characterize features that match the object description features and preference features. The first tag can be used to quickly identify a target object that matches the target features among multiple objects. For example, the first tag can be male, aged 30-40 years old, married, likes car models, likes to use coupons to buy products, etc.

[0082] The second data may be descriptive information of the entity. The descriptive information may be basic information of the entity, such as the type and selling point of the product, the name of the merchant, the products sold, honors, etc.

[0083] The second tag can be used to characterize features that match the entity features and pattern features of the entity. The second tag can be used to quickly obtain the features of the entity. For example, the second tag can be brand A, stationery, coupons can be issued when purchasing the product, free shipping can be obtained when purchasing the product, etc.

[0084] The object set may be a set of objects with a first label and entities with a second label, and the objects in the object set have different degrees of interest in the entities with the second label. For example, users who are interested in coupons for purchasing stationery include user 1, user 2, user 3, and user 4. Users who are interested in free shipping for purchasing stationery include user 1, user 2, and user 5.

[0085] In some embodiments, the first label of an object can be updated periodically (for example, once a month), and the object set corresponding to the first label and the second label can be updated at the same time. Correlation features can be mined based on the first label and the second label of the object in different periods, so that the information recommendation results are more accurate.

[0086] Exemplarily, the description information and behavior information of multiple users and the description information and activity patterns of multiple commodities are analyzed to obtain each user's interest in different commodities and each user's interest in different activity patterns of the same commodity. The corresponding objects for different commodities combined with different activity patterns can be obtained. The object set can be a set of objects with a first label and a second label. For example, object set 1: Users interested in purchasing stationery with coupons include user 1, user 2, user 3, and user 4. User 1, user 2, user 3, and user 4 have the following characteristics in common: 35-40 years old women with children. The first label: 35-40 years old women with children. The second label: interested in purchasing stationery with coupons.

[0087] The target entity is stationery merchant A, and the configuration information is: coupons are issued to users whose interest in stationery products is greater than 10%. In the process of information recommendation, the target feature can be determined as: stationery merchants that issue coupons based on the target entity and the configuration information. Based on multiple object sets (with a first label and a second label). Matching is performed based on the target feature and the second label. When the target feature and the second label are consistent, object set 1 is obtained, and object set 1 is used as the candidate object set. User 1 (interest 50%), user 2 (interest 60%), user 3 (interest 60%), and user 4 (interest 5%) are respectively candidate objects. The candidate objects in object set 1 are screened according to the interest threshold of 10%, and the target object set (user 1 (interest 50%), user 2 (interest 60%), user 3 (interest 60%)) is obtained.

[0088] It is understandable that determining the first label based on the description information and behavior information of the object can more deeply and accurately dig out the object's interest in different entities. By using the first label to more accurately characterize different objects and using the second label to more accurately characterize different entities, it is possible to accurately dig out objects that are interested in the same type of entity. This improves the applicability of the object set in different application scenarios.

[0089] According to an embodiment of the present disclosure, multiple object sets are determined based on a first label and a second label, including: performing a cross-analysis on the first label and the second label to determine the degree of association between the first label and the second label; when the degree of association is greater than a preset threshold, determining an association feature between the first label and the second label; and determining multiple object sets based on the association feature, the first label and the second label.

[0090] Exemplarily, the association feature may be a mapping feature formed by an object having a first label to an entity having a second label. The degree of association may characterize the degree of association between the object having the first label and the entity having the second label. The degree of association between the first label and the second label in the association feature is greater than a preset threshold. For the association feature, an object set corresponding to the association feature may be constructed.

[0091] For example, the correlation between 25-30-year-old women and stationery products is 10, the correlation between 35-40-year-old women with children and stationery products is 50, and the preset correlation threshold is 40. Then, 35-40-year-old women with children and stationery products are determined as the correlation feature, and for 35-40-year-old women with children (first label) and stationery products (second label), 35-40-year-old female users with children can be taken as the objects of the corresponding object set.

[0092] It can be understood that in the process of building an object set, understanding the association between objects and entities can help to more accurately screen out objects of interest to the entity.

[0093] According to an embodiment of the present disclosure, each object set in the plurality of object sets includes at least one object associated with the same entity and the interest level of the at least one object in the same entity.

[0094] The interest level is obtained by the following operation: inputting the associated feature, the first label and the second label into the interest level determination model to obtain the interest level of at least one object with the first label for at least one entity with the second label.

[0095] The object set includes multiple objects, and the multiple objects have different interest levels for the same entity. The object set also has a mapping relationship between the first label and the second label. Here, the mapping relationship between the first label and the second label and the interest level can be added to the object set as additional information.

[0096] The interest of an object in different entities can be predicted by an interest determination model.

[0097] The associated feature, the first data, the second data, the first label and the second label may be simultaneously input into the interest determination model to generate the interest of different objects with the first label to the entity with the second label.

[0098] In one example, 35-40-year-old women with children (first label) - stationery products (second label). The users corresponding to the first label are user 1, user 2, user 3, and user 4. The associated features: 35-40-year-old women with children - stationery products, the first data and the second data of users 1-4 are input into the interest determination model. The interest of each user in stationery products is obtained respectively. For example, user 1 (interest 50%), user 2 (interest 60%), user 3 (interest 60%), user 4 (interest 5%).

[0099] According to the embodiments of the present disclosure, the efficiency of predicting the interest of an object to an entity can be improved by using an interest determination model. At the same time, combined with the association features, the association between the object and the entity can be understood, thereby improving the accuracy of interest prediction.

[0100] In the embodiment of the present disclosure, the plurality of object sets 304 may be determined in advance based on different objects and different entities. Figure 4 Take as an example to illustrate the process of determining the object set.

[0101] Figure 4 The figure schematically shows a process of determining an object set according to an embodiment of the present disclosure.

[0102] like Figure 4 As shown, first data 401 of multiple objects and second data 402 of multiple entities are obtained. According to the first data 401, one or more first tags 403 for each object are generated. According to the second data 402, one or more second tags 404 for each entity are generated. The first tag 403 and the second tag 404 are cross-analyzed. When the feature correlation between the first tag 403 and the second tag 404 is greater than a preset threshold, the first tag 403 and the second tag 404 are determined to be associated features 405. The associated features 405, the first tag 401 and the second tag 402 are input into the interest determination model to generate the interest of the object with the first tag 403 for the entity with the second tag 402, and obtain the object set 304 about the first tag 403 and the second tag 404. Among them, the interest of multiple objects with the first tag 403 in the object set 304 for the second tag 404 is used as the subsidiary information of the object set 304. Finally, the object set 304 of the interest of different objects with respect to different entities is obtained.

[0103] According to an embodiment of the present disclosure, at least one first label is generated for each object based on the first data, including at least one of the following: based on the first data, an attribute label and a first model label of the object are generated, the attribute label includes at least one descriptive feature of the object, and the first model label includes a model type and model parameters used to generate the attribute label; based on the first data, a statistical label and a second model label of the object are generated, the statistical label includes at least one statistical feature of the object, and the second model label includes a model type and model parameters used to generate the statistical label; based on the first data, a preference label and a third model label of the object are generated, the preference label includes the object's preference features for different entities, and the third model label includes a model type and model parameters used to generate the preference features.

[0104] In one example, a property label and a first model label of the object are generated according to the first data.

[0105] Exemplarily, attribute tags can be used to mark users based on the terminal device used by the user. Attribute tags can be attributes of the user's description information and can be directly extracted from the user's description information data. For example, user gender, age, region, etc., mobile phone brand, operating system and other device information, attribute tags can be used in target object screening, and can also be used as input parameters of preference tags.

[0106] In the process of obtaining different attribute labels of the user, the first model can be used. The first data of the user is input into the first model to generate the attribute label of the user. At the same time, the model parameters of the first model can be stored. In the case of subsequently generating attribute labels according to the user data, the first model with corresponding parameters can be used to improve the efficiency of label generation.

[0107] In another example, a statistical label and a second model label of the object are generated based on the first data.

[0108] For example, the statistical tags can be combined with the user's first data for statistical grouping to reflect the characteristics of the user data. For example, through the user's access behavior, the user's preference for e-commerce product categories, user activity and other tags can be judged. Statistical tags can be used in personalized recommendations of information flow, activation of new and old users, etc.

[0109] In the process of obtaining different statistical labels of the user, the second model can be used. The first data of the user is input into the second model to generate the statistical label of the user. At the same time, the model parameters of the second model can be stored. In the case of subsequent statistical label generation based on the user data, the second model with corresponding parameters can be used to improve the efficiency of label generation.

[0110] In yet another example, a preference label and a third model label of the object are generated based on the first data.

[0111] For example, the prediction tag can be a feature that predicts the user's behavior preferences and tendencies based on the user's first data. The prediction tag mainly uses a deep learning algorithm to calculate the user's consumption intention, attention intention, category preference and other tags. For example, if a user has recently browsed a lot of winter down jackets and has a medium-to-high income, the price acceptance range and brand type for down jackets can be predicted.

[0112] In the process of generating prediction labels by combining the user's first data, statistical labels, and attribute labels, different third models can be used for different types of prediction labels, such as machine learning models such as logistic regression, decision tree, and principal component analysis. For example, for the prediction label of the user's product consumption intention, a decision tree model can be used. For the user's price sensitivity, a logistic regression model can be used.

[0113] In yet another example, based on the first data, an attribute label, a first model label, a statistical label, a second model label, a preference label, and a third model label of the object are generated.

[0114] It should be noted that any type of attribute label, statistical label and preference label can be selected according to actual application requirements. This embodiment of the present disclosure does not limit this. When selecting different types of labels, the model label corresponding to the type of label is also required to be used as the generated result.

[0115] It is understandable that attribute tags ensure that the information recommendation system can provide highly relevant recommendations based on the specific characteristics of the object; statistical tags make the recommended content more in line with the actual needs of the object by analyzing the behavior data of the object; and predictive tags can explore the user's potential needs in advance and make forward-looking recommendations. Therefore, the use of user attribute tags, statistical tags, and predictive tags can enhance the intelligence, personalization, and accuracy of information recommendations, making information recommendations more in line with user needs, improving user experience, and promoting the platform's business growth.

[0116] In the embodiments of the present disclosure, information recommendation can be made through analysis at different levels. Figure 5 Take as an example to illustrate the application process of information recommendation.

[0117] Figure 5 The application diagram of the information recommendation method according to the embodiment of the present disclosure is schematically shown.

[0118] like Figure 5As shown. In the operating environment 501, the user generates access behaviors through different applications in the terminal to form the user's original information, and the first original information is stored in the server as log data. In the embedded log 502, the log management system can analyze the log data in the server by means of embedded logs to obtain the user's original information. In the database 503, the original information can be stored in different types of databases according to the needs of different data applications. At the same time, the basic information of different entities can also be recorded in the database for storage. In the data warehouse 504, the original information of the user and the basic information of different entities obtained from the database are input into the data warehouse. The different data layers in the data warehouse clean, count, aggregate, etc. the original information and basic information to generate the user's data for analysis and decision-making: the first data and the entity's data: the second data. In the object portrait 505, according to the first data of different users, through different analysis models, multiple portrait labels for each user are analyzed. In the entity configuration 506, the object set of the user about different entities is determined in advance according to the labels of multiple users. For details, please refer to the above Figure 4 According to the configuration information of the target entity, determine the target object associated with the target entity, and push the target entity to the target object. Figure 3 In the dashboard 507, the interaction between the user and the target entity can be analyzed to generate different dashboards, which can be used to display the recommendation results of the target entity for the user in multiple dimensions.

[0119] In one example, when the embedded point log data requires strict data integrity, the first database can be used: a relational database, such as a MySQL database. If the user data inventory is large, the second database can be used: a distributed file system. The third database can also be used: a time series database to store user data that requires time series data.

[0120] In one example, the data warehouse may include a data application layer, a data dimension layer, a data service layer, an intermediate data layer, a data access layer, and a basic data layer.

[0121] The data dimension layer is used to store data of different dimensions of users or entities. For example, the dimension table of the live broadcast room, the dimension table of the anchor information, etc. The data access layer is used to obtain the original data of users from different business systems, usually including the log system, order system, logistics system, comment system, etc. The basic data layer is used to clean and convert the data of the data access layer to remove redundant, erroneous or useless data to ensure the accuracy and neatness of the data. Among them, the data foundation layer does not involve complex logic, model algorithms and other data. The intermediate data layer is used to aggregate the data of the basic data layer. For example, by introducing risk control models, cleaning rules, etc., more in-depth analysis, feature extraction and data aggregation can be carried out. The data service layer is used to build data tables according to business themes based on specific business needs. The data application layer is used to extract data from the data warehouse according to business needs, perform necessary logical operations and processing, and generate visual reports or analysis results.

[0122] In one example, for user tags, a first model may be used to analyze the first data to obtain attribute tags, a second model may be used to analyze the first data to obtain statistical tags, and a third model may be used to analyze the first data to obtain preference tags.

[0123] In one example, the results of information recommendation can be visualized and analyzed. From basic user behavior to order completion, different scenarios have different conversion paths. In order to facilitate indicator monitoring and business analysis, a merchant store detail dashboard (first dashboard 601), a user market distribution dashboard (second dashboard 602), and a dashboard for the effects of different activity modes (third dashboard) can be constructed.

[0124] In an embodiment of the present disclosure, a dashboard may be generated based on the result of information recommendation. Figure 6 Take the following as an example to illustrate the display result of the first dashboard.

[0125] Figure 6 A schematic diagram of a first signboard according to an embodiment of the present disclosure is schematically shown.

[0126] like Figure 6 As shown, the support provided by an e-commerce platform to an e-commerce merchant is used as an example for explanation. After the merchant's products are recommended to the user, the user may purchase the products. Therefore, a first dashboard 601 can be generated for the result of the information recommendation. In the first dashboard 601, data analysis can be performed on the user's purchase of products, so that the merchant can view the corresponding configuration information to perform information recommendation. For example, the first dashboard can display the new customer transaction amount, the number of new customer transactions, the new customer unit price, the new customer conversion rate, the new customer conversion analysis, the customer data trend and the product data trend.

[0127] According to the embodiment of the present disclosure, the configuration information is determined by using the following operations: obtaining the working parameters of the target entity; and determining the configuration information corresponding to the target entity according to the working parameters.

[0128] Exemplarily, the working parameters may be different stages of the entity's work. For example, the working parameters may include: primary stage and non-primary stage. For example, it may be the first time for a merchant to enter an e-commerce platform. It may also be the first time for a host to broadcast live.

[0129] Different strategies can be used to determine the configuration information of an entity at different stages. For example, for merchants in non-primary stages, the activity mode can be determined based on the merchant's historical data. For merchants in the primary stage, the configuration information of the merchant can be determined based on the information recommendation results of successful merchants similar to the merchant. If stationery merchant A enters the e-commerce platform for the first time. Stationery merchant B is a merchant with good sales results among the stationery merchants on the e-commerce platform, then stationery merchant A can use configuration information similar to that of stationery merchant B for promotion and application.

[0130] According to the disclosed embodiment, different configuration strategies are adopted according to different working parameters of the target entity to determine the configuration information of the target entity. By flexibly adjusting the configuration information at different stages, the entity can not only better cope with market changes, but also maximize resource utilization efficiency while maintaining user loyalty.

[0131] According to an embodiment of the present disclosure, configuration information for a target entity is determined based on working parameters, including: when the working parameters do not meet preset conditions, obtaining an associated entity, the associated entity having the same second tag as the target entity, and the associated entity meeting the preset conditions, the preset conditions indicating that the entity has historical data; and determining configuration information for the target entity based on the historical configuration information of the associated entity.

[0132] The preset condition may be whether the entity is in a non-primary stage. For example, if stationery merchant A enters the e-commerce platform for the first time, then stationery merchant A does not meet the preset condition.

[0133] The description features of the associated entity and the target entity are the same. For example, stationery merchant A and stationery merchant B have the same description feature: stationery.

[0134] In one example, if stationery merchant A is entering the e-commerce platform for the first time, and stationery merchant B is a merchant that has entered the e-commerce platform in the past and has good sales results among the stationery merchants on the e-commerce platform. After analyzing the historical data of stationery merchant B, it is found that the activity mode of stationery merchant B uses coupons for better sales results. Then stationery merchant A can use configuration information similar to that of stationery merchant B: use coupons, identify target users and promote.

[0135] In another example, merchants use the platform for live broadcasting. The live broadcasting architecture pushes targeted crowd data based on the configured support strategy, which can be intelligent and accurate. The platform can configure coupons, which can be accurately distributed to all selected users. When merchants cold-start live broadcasting, the merchant's live broadcast room is recommended and has corresponding coupons, the user conversion rate will be very high, greatly improving the efficiency of traffic use.

[0136] It is understandable that, by setting preset conditions, for target entities that do not meet the preset conditions, the configuration information of the target entity can be determined based on the historical configuration information of the associated entities with the same second label as the target entity. This can not only help the target entity to quickly screen out target objects, but also improve the accuracy of the target entity's recommendations on target objects.

[0137] According to an embodiment of the present disclosure, the method also includes: obtaining weight values ​​corresponding to each activity mode in the configuration information; obtaining interaction data and conversion data for each activity mode, the interaction data including the number of interactions between the object and the target entity through the corresponding activity mode, and the conversion data including the number of objects participating in the activity mode; determining the contribution value of each activity mode to the conversion data based on the interaction data, conversion data and weight values; and updating the configuration information of the target entity based on the contribution value.

[0138] The weight value can be the importance of different activity modes. For example, for merchant A, three activity modes are configured: activity mode 1: playing merchant A's advertisement on the application startup interface, corresponding to the first weight value: 0.5; activity mode 2: promoting merchant A through media bloggers, corresponding to the second weight value: 0.3; activity mode 3: issuing coupons, corresponding to the third weight value: 0.2. The weight value can be determined by expert experience or generated according to the model.

[0139] Interaction data can be the number of interactions between a user and a target entity through a certain activity mode. For example, a user views merchant A by clicking on an advertisement in activity mode 1.

[0140] Conversion data can be the number of users who participate in an activity mode through a certain activity mode. Participation in an activity mode can be the object meeting the conversion index. The conversion index can be a key indicator for measuring the effectiveness of a merchant, such as sales, number of orders, number of user registrations, purchase frequency, etc. For example, if a user views a merchant by clicking on an advertisement in activity mode 1 and purchases a product from merchant A, the user is counted in the conversion data.

[0141] When the conversion data is ordering behavior, the interaction data may include multiple interaction modes: at least one of browsing behavior, collection behavior, adding to shopping cart behavior, consulting behavior, etc.

[0142] The contribution value may be the contribution degree of different activity modes to the conversion effect of the target entity.

[0143] Based on the weight values ​​of different interaction modes in different activity modes, interaction data and conversion data, the contribution value of each activity mode to the conversion effect of the target entity can be calculated using the weighted attribution method.

[0144] According to the embodiments of the present disclosure, by analyzing different activity patterns in the configuration information and determining the impact of each activity pattern on the conversion of the number of target entity users, merchants can analyze the results of the activity patterns and obtain a more effective target entity information recommendation method for objects.

[0145] In an example, in e-commerce applications, for a population package with good conversion effects and a small number of target objects, the descriptive features of the target entity can be used as public population data with the descriptive features to serve different merchants in the same industry. Based on the different features of different merchants in the same industry, attribution analysis can be performed and the population can be continuously iterated. Finally, a universal population package for the industry is formed to further improve the efficiency of industry support.

[0146] In an embodiment of the present disclosure, the second dashboard can be generated based on the results of attribution analysis of each activity mode. Figure 7 For example, a second dashboard 602 showing the attribution analysis result.

[0147] Figure 7 A schematic diagram of a second dashboard 602 according to an embodiment of the present disclosure is schematically shown.

[0148] like Figure 7 As shown, the interaction data, conversion data, and attribution values ​​corresponding to different activity modes (eg, activity mode 1, activity mode 2, and activity mode 3) can be displayed to the merchant in the form of a data table to form a second dashboard 602 .

[0149] It can be understood that by setting up the second signboard, the effects of different activity modes on user conversion can be intuitively displayed to merchants.

[0150] It can be understood that the method of the present disclosure is described above, and the device of the present disclosure will be described below.

[0151] Figure 8 is a block diagram of an information recommendation device according to an embodiment of the present disclosure.

[0152] like Figure 8 As shown, the device 700 may include a first determination module 710 , a second determination module 720 and a push module 730 .

[0153] The first determination module 710 is used to determine entity features for the target entity and mode features of the corresponding activity mode according to configuration information of the target entity, wherein the configuration information represents at least one activity mode of the target entity with respect to the object.

[0154] The second determination module 720 is used to determine a target object set corresponding to the target entity according to the interest of multiple objects with respect to the target entity, wherein the target object set includes at least one target object, and the interest of the target object with respect to the target entity is greater than an interest threshold.

[0155] The push module 730 is configured to push a target entity and a corresponding activity mode to at least one target object in the target object set.

[0156] According to an embodiment of the present disclosure, the second determination module includes a first acquisition submodule, a first determination submodule and a second determination submodule.

[0157] The first acquisition submodule is used to acquire at least one target feature, where the target feature includes an entity feature and a pattern feature.

[0158] The first determination submodule is used to determine, according to at least one target feature, from multiple object sets a candidate object set associated with the at least one target feature, wherein the candidate object set includes at least one candidate object.

[0159] The second determination submodule is used to determine the candidate object as the target object when the interest of the candidate object to the target entity is greater than the interest threshold.

[0160] According to the embodiment of the present disclosure, the device may further include a first acquisition module, a first generation module, a second generation module and a third determination module.

[0161] The first acquisition module is used to acquire first data of multiple objects and second data of multiple entities, the first data represents description information and behavior information of the objects, and the second data represents description information of the entities.

[0162] The first generating module is used to generate at least one first label for each object according to the first data, wherein the first label includes at least one of a descriptive feature and a preference feature of the object.

[0163] The second generating module is used to generate at least one second label for each entity according to the second data, and the second label includes entity features and pattern features of the entity.

[0164] The third determination module is used to determine multiple object sets based on the first label and the second label, and the object sets represent the interest of the object with the first label to the entity with the second label.

[0165] According to an embodiment of the present disclosure, the third determination module may include a third determination submodule, a fourth determination submodule and a fifth determination submodule.

[0166] The third determination submodule is used to perform cross analysis on the first tag and the second tag to determine the correlation between the first tag and the second tag.

[0167] The fourth determination submodule is used to determine the association feature between the first tag and the second tag when the association degree is greater than a preset threshold.

[0168] The fifth determination submodule is used to determine multiple object sets according to the associated features, the first label and the second label.

[0169] According to an embodiment of the present disclosure, each object set in the plurality of object sets includes at least one object associated with the same entity and the interest level of the at least one object in the same entity, and the device further includes: a processing module.

[0170] The processing module is used to input the associated features, the first label and the second label into the interest determination model to obtain the interest of at least one object with the first label to at least one entity with the second label.

[0171] According to an embodiment of the present disclosure, the first generation module further includes at least one of the following: a first generation submodule, a second generation submodule, or a third generation submodule.

[0172] The first generating submodule is used to generate an attribute label and a first model label of the object according to the first data, wherein the attribute label includes at least one descriptive feature of the object, and the first model label includes a model type and model parameters used to generate the attribute label.

[0173] The second generating submodule is used to generate a statistical label and a second model label of the object according to the first data, wherein the statistical label includes at least one statistical feature of the object, and the second model label includes a model type and model parameters used to generate the statistical label.

[0174] The third generating submodule is used to generate a preference label and a third model label of the object according to the first data, wherein the preference label includes the preference features of the object for different entities, and the third model label includes the model type and model parameters used to generate the preference features.

[0175] According to the embodiment of the present disclosure, the device further includes: a second acquisition module and a fourth determination module.

[0176] The second acquisition module is used to acquire the working parameters of the target entity.

[0177] The fourth determination module is used to determine the configuration information corresponding to the target entity according to the working parameters.

[0178] According to the embodiment of the present disclosure, the fourth determination module includes: a second acquisition submodule and a sixth determination submodule.

[0179] The second acquisition submodule is used to acquire the associated entity when the working parameter does not meet the preset condition, the associated entity has the same second label as the target entity, and the associated entity meets the preset condition, and the preset condition represents that the entity has historical data.

[0180] The sixth determination submodule is used to determine the configuration information for the target entity according to the historical configuration information of the associated entity.

[0181] According to the embodiment of the present disclosure, the device further includes: a third acquisition module, a fourth acquisition module, a fifth determination module and an update module.

[0182] The third acquisition module is used to obtain the weight value corresponding to each activity mode in the configuration information.

[0183] The fourth acquisition module is used to acquire the interaction data and conversion data of each activity mode, wherein the interaction data includes the number of interactions between the object and the target entity through the corresponding activity mode, and the conversion data includes the number of the object participating in the activity mode.

[0184] The fifth determination module is used to determine the contribution value of each activity mode to the conversion data according to the interaction data, the conversion data and the weight value.

[0185] The update module is used to update the configuration information of the target entity according to the contribution value.

[0186] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0187] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0188] In some embodiments, an electronic device includes: at least one processor; and a memory connected to the at least one processor in communication. The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform a method provided in accordance with the present disclosure (e.g., the above-mentioned method 200).

[0189] In some embodiments, a non-transitory computer-readable storage medium stores computer instructions, where the computer instructions are used to enable a computer to execute a method provided according to the present disclosure (eg, the above-mentioned method 200).

[0190] In some embodiments, a computer program product includes a computer program, and when the computer program is executed by a processor, the computer program implements the method provided according to the present disclosure (such as the above method 200). Fig. 9 Provide further explanation.

[0191] Fig. 9 1 is a block diagram of an electronic device to which an information recommendation method can be applied according to an embodiment of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0192] like Fig. 9 As shown, the device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 to a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 500 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0193] A number of components in the device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the device 800 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0194] The computing unit 801 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as the information recommendation method. For example, in some embodiments, the information recommendation method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the information recommendation method described above may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to execute the information recommendation method in any other appropriate manner (eg, by means of firmware).

[0195] Various embodiments of the systems and techniques described above herein may be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard parts (ASSPs), system on chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0196] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0197] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories, read-only memories, erasable programmable read-only memories (EPROM) or flash memories, optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0198] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a cathode ray tube (CRT) display or a liquid crystal display (LCD)) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0199] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a Local Area Network (LAN), a Wide Area Network (WAN), and the Internet.

[0200] A computer system may include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other.

[0201] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0202] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. An information recommendation method, comprising: Determining entity features and mode features of corresponding activity modes for the target entity according to configuration information of the target entity, the configuration information characterizing at least one activity mode of the target entity with respect to an object; Determining a target object set corresponding to the target entity according to the interest of a plurality of objects with respect to the target entity, the target object set including at least one target object, the interest of the target object with respect to the target entity being greater than an interest threshold; The target entity and the corresponding activity mode are pushed to at least one target object in the target object set.

2. The method according to claim 1, wherein: Determining a target object set corresponding to the target entity according to the interest of multiple objects in the target entity includes: Acquire at least one target feature, wherein the target feature includes the entity feature and the pattern feature; According to the at least one target feature, determining a candidate object set associated with the at least one target feature from a plurality of object sets, the candidate object set including at least one candidate object; In the case that the interest of the candidate object to the target entity is greater than the interest threshold, the candidate object is determined to be the target object.

3. The method according to claim 2, wherein: The plurality of object sets are determined by the following operations: Acquire first data of a plurality of objects and second data of a plurality of entities, wherein the first data represents description information and behavior information of the objects, and the second data represents description information of the entities; generating at least one first label for each of the objects according to the first data, the first label comprising at least one of a descriptive feature and a preference feature of the object; generating at least one second label for each of the entities according to the second data, the second label comprising entity features and pattern features of the entity; A plurality of object sets are determined according to the first label and the second label, wherein the object sets represent the interest of the objects with the first label in the entity with the second label.

4. The method according to claim 3, wherein: Determining a plurality of object sets according to the first label and the second label includes: Performing a cross analysis on the first tag and the second tag to determine a correlation degree between the first tag and the second tag; When the degree of association is greater than a preset threshold, determining an association feature between the first tag and the second tag; A plurality of object sets are determined according to the association feature, the first label, and the second label.

5. The method according to claim 4, wherein: Each object set in the plurality of object sets includes at least one object associated with the same entity and the interest level of the at least one object with respect to the same entity; the interest level is obtained by: The associated features, the first label and the second label are input into an interest determination model to obtain the interest of at least one object with the first label to at least one entity with the second label.

6. The method according to claim 3, wherein: Generating at least one first label for each of the objects according to the first data includes at least one of the following: Generate an attribute label and a first model label of the object according to the first data, wherein the attribute label includes at least one descriptive feature of the object, and the first model label includes a model type and model parameters used to generate the attribute label; Generate a statistical label and a second model label of the object according to the first data, wherein the statistical label includes at least one statistical feature of the object, and the second model label includes a model type and model parameters used to generate the statistical label; A preference label and a third model label of the object are generated according to the first data, wherein the preference label includes the preference characteristics of the object for different entities, and the third model label includes the model type and model parameters used to generate the preference characteristics.

7. The method according to claim 3, wherein: The configuration information is determined using the following operations: Obtaining working parameters of the target entity; According to the working parameters, configuration information corresponding to the target entity is determined.

8. The method according to claim 7, wherein: The determining, according to the working parameter, configuration information for the target entity comprises: When the working parameter does not satisfy the preset condition, obtaining an associated entity, wherein the associated entity has the same second tag as the target entity and the associated entity satisfies the preset condition, wherein the preset condition indicates that the entity has historical data; Configuration information for the target entity is determined based on the historical configuration information of the associated entity.

9. The method according to claim 1, further comprising: Obtaining a weight value corresponding to each of the activity modes in the configuration information; Acquire interaction data and conversion data of each of the activity modes, wherein the interaction data includes the number of interactions between the object and the target entity through the corresponding activity mode, and the conversion data includes the number of objects participating in the activity mode; Determining a contribution value of each of the activity patterns to the conversion data according to the interaction data, the conversion data and the weight value; According to the contribution value, the configuration information of the target entity is updated.

10. An information recommendation device, comprising: A first determination module, configured to determine entity features for the target entity and mode features of corresponding activity modes according to configuration information of the target entity, wherein the configuration information represents at least one activity mode of the target entity with respect to an object; A second determination module is used to determine a target object set corresponding to the target entity according to the interest of multiple objects with respect to the target entity, wherein the target object set includes at least one target object, and the interest of the target object with respect to the target entity is greater than an interest threshold; A pushing module is used to push the target entity and the corresponding activity mode to at least one target object in the target object set.

11. The device according to claim 10, wherein: The second determining module comprises: A first acquisition submodule is used to acquire at least one target feature, wherein the target feature includes the entity feature and the pattern feature; A first determination submodule is configured to determine, according to the at least one target feature, a candidate object set associated with the at least one target feature from a plurality of object sets, wherein the candidate object set includes at least one candidate object; The second determination submodule is configured to determine the candidate object as the target object if the interest of the candidate object to the target entity is greater than the interest threshold.

12. The device according to claim 11, wherein The device also includes: A first acquisition module, used to acquire first data of multiple objects and second data of multiple entities, wherein the first data represents description information and behavior information of the objects, and the second data represents description information of the entities; A first generating module, configured to generate at least one first label for each of the objects according to the first data, wherein the first label includes at least one of a descriptive feature and a preference feature of the object; a second generating module, configured to generate at least one second label for each of the entities according to the second data, wherein the second label includes entity features and pattern features of the entity; The third determination module is used to determine a plurality of object sets according to the first label and the second label, wherein the object sets represent the interest of the object with the first label in the entity with the second label.

13. The device according to claim 12, wherein: The third determination module comprises: A third determination submodule, configured to perform a cross analysis on the first label and the second label to determine a correlation between the first label and the second label; A fourth determination submodule, configured to determine an association feature between the first tag and the second tag when the association degree is greater than a preset threshold; The fifth determining submodule is used to determine multiple object sets according to the associated features, the first label and the second label.

14. The device according to claim 13, wherein: Each object set in the plurality of object sets includes at least one object associated with a same entity and the interest level of the at least one object in the same entity, and the apparatus further includes: The processing module is used to input the associated features, the first label and the second label into an interest determination model to obtain the interest of at least one object with the first label to at least one entity with the second label.

15. The device according to claim 12, wherein: The first generation module further includes at least one of the following: A first generating submodule, configured to generate an attribute label and a first model label of the object according to the first data, wherein the attribute label includes at least one descriptive feature of the object, and the first model label includes a model type and model parameters used to generate the attribute label; A second generating submodule is used to generate a statistical label and a second model label of the object according to the first data, wherein the statistical label includes at least one statistical feature of the object, and the second model label includes a model type and model parameters used to generate the statistical label; The third generating submodule is used to generate a preference label and a third model label of the object according to the first data, wherein the preference label includes the preference characteristics of the object for different entities, and the third model label includes a model type and model parameters used to generate the preference characteristics.

16. The device according to claim 12, wherein: The device also includes: A second acquisition module, used to acquire the working parameters of the target entity; The fourth determination module is used to determine the configuration information corresponding to the target entity according to the working parameters.

17. The device according to claim 16, wherein: The fourth determination module comprises: A second acquisition submodule is used to acquire an associated entity when the working parameter does not meet a preset condition, the associated entity having the same second tag as the target entity and the associated entity meeting a preset condition, wherein the preset condition indicates that the entity has historical data; The sixth determining submodule is used to determine the configuration information for the target entity according to the historical configuration information of the associated entity.

18. The device according to claim 10, wherein: The device also includes: A third acquisition module is used to obtain a weight value corresponding to each of the activity modes in the configuration information; a fourth acquisition module, configured to acquire interaction data and conversion data of each of the activity modes, wherein the interaction data includes the number of interactions between the object and the target entity through the corresponding activity mode, and the conversion data includes the number of objects participating in the activity mode; a fifth determining module, configured to determine a contribution value of each of the activity patterns to the conversion data according to the interaction data, the conversion data and the weight value; An updating module is used to update the configuration information of the target entity according to the contribution value.

19. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 9.

20. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 9.

21. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 9.

Images