Method, device and equipment for content recommendation, medium and program product

Abstract

According to the embodiment of the invention, a content recommendation method and device, equipment, a medium and a program product are provided. The method comprises the steps that in response to a received supply request for target recommended content, category information corresponding to the target recommended content is determined, and the category information indicates a target category into which the target recommended content is classified and / or at least one category related to the target category; obtaining historical behavior data associated with the category indicated by the category information, wherein the historical behavior data indicates at least one behavior executed by the audience group on the recommended content when the recommended content of the corresponding category is provided for the audience group; and based on the historical behavior data, determining a target audience group corresponding to the target recommendation content so as to provide the target recommendation content for at least part of audiences in the target audience group.

Description

[0001]The technical field of the method, device, equipment, medium and program product for content recommendation relates to the computer field, and particularly to a method, device, electronic equipment, computer readable storage medium and computer program product for content recommendation. Background art The Internet provides access to a variety of resources. For example, various applications, goods, audio and video content, etc. can be accessed through the Internet. In addition, content delivery and service promotion through the Internet have become a new form of information dissemination and are widely used. An advertising system supports displaying recommended content or services to users in different advertising display opportunities, enabling users to browse, obtain corresponding services, etc. according to needs. In a first aspect of the present disclosure, a method for content recommendation is provided. The method comprises: in response to receiving a supply request for target recommended content, determining category information corresponding to the target recommended content, the category information indicating a target category to which the target recommended content is classified and / or at least one category related to the target category; obtaining historical behavior data associated with the category indicated by the category information, the historical behavior data indicating at least one behavior performed by an audience group on recommended content of the corresponding category when the recommended content is provided to the audience group; and determining a target audience group corresponding to the target recommended content based on the historical behavior data, for providing the target recommended content to at least part of the target audience group. In a second aspect of the present disclosure, a device for content recommendation is provided. The device comprises: a category determination module configured to determine, in response to receiving a supply request for target recommended content, category information corresponding to the target recommended content, the category information indicating a target category to which the target recommended content is classified and / or at least one category related to the target category; a data obtaining module configured to obtain historical behavior data associated with the category indicated by the category information, the historical behavior data indicating at least one behavior performed by an audience group on recommended content of the corresponding category when the recommended content is provided to the audience group; and an audience determination module configured to determine, based on the historical behavior data, a target audience group corresponding to the target recommended content, for providing the target recommended content to at least part of the target audience group. In a third aspect of the present disclosure, an electronic device is provided. The device comprises at least one processing unit; and at least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit. The instructions, when executed by the at least one processing unit, cause the device to perform the method of the first aspect. In a fourth aspect of the present disclosure, a computer readable storage medium is provided. The medium has computer executable instructions stored thereon, which, when executed by a processor, implement the method of the first aspect.In a fifth aspect of the disclosure, a computer program product is provided. The computer program product is tangibly stored in a computer storage medium and includes computer-executable instructions that, when executed by a device, cause the device to perform the method of the first aspect. It should be understood that all statements made herein are intended to be non-limiting and are intended to be construed as encompassing all the features and benefits described herein, and all equivalent features and benefits. One skilled in the art will readily recognize from the disclosure herein, that optional or desired steps from the described methods can be incorporated into the methods and / or devices disclosed herein, and that the steps need not be performed in any particular order unless specified. It will also be readily apparent that one or more steps of a method can be performed by more than one component of a device. Furthermore, it will be readily apparent to those skilled in the art that the steps of the methods described herein are intended to be performed by any suitable means desired or otherwise indicated. These and other modifications can be made to the disclosure in light of the foregoing description. The terms “coupled” and “connected,” along with derivatives thereof, can be used. It should be understood that these terms are not intended as synonyms for each other. Rather, these terms can be used in certain i contexts to convey that there is a connection between two elements or modules, electrical coupling, while in other contexts to convey that there is a physical or electrical connection between two elements or modules, mechanical coupling. Likewise, the term “coupled or connected” can be used. It should be understood that these terms are not intended as synonyms for one another. Rather, these terms can be used in certain contexts to convey that there is a connection between two elements or modules, electrical coupling, while in other contexts to convey that there is a physical or electrical connection between two elements or modules, mechanical coupling. Other definitions can be found elsewhere in the disclosure.For example, in response to receiving an active request of a user, a prompt information is sent to the user to explicitly prompt the user that the operation requested to be performed by the user will require obtaining and using personal information of the user, so that the user can autonomously select whether to provide the personal information to the software or hardware such as an electronic device, an application, a server or a storage medium performing the operation of the technical solution of the present disclosure according to the prompt information. As an optional but non-limiting implementation manner, in response to receiving an active request of a user, the manner of sending a prompt information to the user may, for example, be a pop-up window manner, and the prompt information may be presented in the pop-up window in the form of text. In addition, the pop-up window may also carry a selection control for the user to select "agree" or "disagree" to provide personal information to the electronic device. It can be understood that the above notification and obtaining of user authorization process is only illustrative, and does not limit the implementation manner of the present disclosure, and other manners meeting the relevant laws and regulations can also be applied to the implementation manner of the present disclosure. FIG. 1 shows a schematic diagram of an example environment 100 in which embodiments of the present disclosure can be implemented. One or more content providers can use a recommendation management system 150 to manage content to be placed on a content placement platform 110. One or more client devices 130-1, 130-2, 130-3, etc. (collectively or individually referred to as client devices 130 for ease of discussion) are associated with the content placement platform 110 and can access various types of content provided on the content placement platform 110, for example, based on respective audiences 132-1, 132-2, 132-3, etc. (collectively or individually referred to as audiences 132 for ease of discussion). As an example, the content placement platform 110 can be an application, a website, a web page, and other accessible platforms. The client devices 130 can be installed with an application for accessing the content placement platform 110, or can access the content placement platform 110 in a suitable manner. The content placement platform 110 can be configured to provide one or more specific recommended contents (e.g., provided or presented on the client devices 130) related to one or more objects to an audience group based on a corresponding strategy. The recommended contents to be provided may, for example, include one or more recommended contents 122-1, 122-2, … 122-M (collectively or individually referred to as recommended contents 122 for ease of discussion) in a content database 120. In this document, an object mayIn this document, an audience group can include one or more user members, e.g., an audience 132o user member can be any potential consumer of a service, e.g., a user, a group, an organization, an entity, etc. In some embodiments, the content placement platform 110 can provide corresponding recommended content 122 to an audience group based on a request by a recommendation requestor 152-1, 152-2, 152-3, etc. (collectively or individually referred to as “recommendation requestor” 152). OIn the context of ad serving, the service provider is also sometimes referred to as an advertiser. In the environment 100, the client device 130 can be any type of mobile terminal, fixed terminal, or portable terminal including a mobile handset, a tablet computer, a notebook computer, a laptop computer, a netbook computer, a tablet computer, a media computer, a multimedia tablet, a personal navigation device, a personal digital assistant (PDA), an audio / video player, a digital camera / camcorder, a positioning device, a television receiver, a radio broadcast receiver, an e-book device, a game device, or any combination thereof, including accessories and peripherals of such devices, or any combination thereof. In some embodiments, the client device 130 can also be capable of supporting any type of interface for user interaction (such as "wearable" circuitry, etc.). In the environment 100, the content serving platform 110 and / or the recommendation management system 150 can be, for example, various types of computing systems / servers capable of providing computing capabilities, including but not limited to mainframes, edge computing nodes, computing devices in a cloud environment, etc. While shown separately, one or more of the content serving platform 110 and / or the recommendation management system 150 can be combined. It should be appreciated that the components and arrangement of the environment shown in FIG. 1 are merely examples, and a computing system suitable for implementing the example embodiments described in this disclosure can include one or more different components, other components, and / or different arrangements. Generally speaking, a platform can provide corresponding recommended content to a target audience group upon request of a recommendation requester (e.g., an advertiser) to provide information related to the recommended object to the target audience group. The initial stage of the entire process of providing the recommended content can include a targeting stage for indicating or defining the target audience group to be served. In the subsequent stage, it will be determined whether the recommended content is to be provided to a particular audience in the target audience group in one or more content presentation opportunities according to the specific circumstances of the audience and the specific circumstances of the recommended content. In a conventional scheme, the target audience group to be served is usually manually selected by the recommendation requester. For example, the recommendation requester can provide some interest tags, keywords, etc. for filtering the target audience group to which the recommended content is to be served. However, the selection of the recommendation requester can not be accurate, resulting in that the recommended content cannot be provided to a more suitable target audience group, which can affect the conversion result of the recommended content by the user. According to an embodiment of the present disclosure, an optimized content recommendation scheme is provided. According to the scheme, upon receiving a service request for newly arrived target recommended content, the category information corresponding to the target recommended content is determined.The category information indicates a target category to which the target recommendation content is classified, and / or at least one category related to the target category. Historical behavior data associated with the category indicated by the category information is obtained, the historical behavior data indicating at least one behavior performed by the audience group on the recommendation content when the recommendation content of the corresponding category is provided to the audience group. Based on the historical behavior data, a target audience group corresponding to the target recommendation content is determined for providing the target recommendation content to at least part of the target audience group. In this way, the historical behavior data under each category can be obtained as a reference according to the category related to the recommendation content, so as to adapt a more matched audience group for the target recommendation content. Such an adaptive audience group determination process can not only reduce the work complexity of the recommendation requester, but also quickly match an accurate audience group for content recommendation, reduce the trial and error cost of the recommendation requester, and improve the conversion effect of the recommendation content. Some example embodiments of the present disclosure will be described below with reference to the accompanying drawings. FIG. 2 shows a flowchart of an example process 200 for content recommendation according to some embodiments of the present disclosure. The process 200 can be implemented by the recommendation management system 150 shown in FIG. 1, for example. The process 200 will be described below with reference to FIG. 1. In the process 200, at block 210, the recommendation management system 150 detects a provision request for newly arrived target recommendation content. The recommendation management system 150 can interface to one or more recommendation requesters 152, such as an advertiser, for example. The recommendation requester 152 can determine to provide recommendation content related to a particular object to a certain audience group in the content delivery platform 110 as needed. In some embodiments, the recommendation requester 152 can initiate a provision request for a recommendation plan, which can indicate one or more recommendation contents to be provided for a particular object, and can also include provision time, period, and bid information for each provision of the recommendation content, etc. When providing the recommendation content, it is usually desired that as many audience groups as possible can perform a particular behavior or interaction on the recommendation content, which is referred to as conversion of the recommendation content. The specific behavior or interaction observed can be various, which can include but is not limited to clicking, searching, liking, clicking interest, commenting, purchasing, downloading, etc. For a particular recommendation content, a series of behaviors can be performed by a certain audience. For recommendation contents related to different objects, the behaviors of interest or executable behaviors can be different. For example, for recommendation contents corresponding to application products, the conversion behavior can include downloading, and for goods in an online shopping scenario, the conversion behavior can include being added to a shopping cart and purchasing. At block 220, the recommendation management system 150 determines category information corresponding to the target recommendation content.The category information indicates a target category to which the target recommendation content is classified. Alternatively or additionally, the category information further indicates at least one category that is related to the target category. In some embodiments, a classification machine learning model can be utilized to classify the target recommendation content or its related recommendation content, and determine the target category to which the target recommendation content is classified. In some embodiments, the target category to which the target recommendation content is classified can be a category corresponding to an object to which the target recommendation content is to be recommended. For example, the target recommendation content can be an advertisement content about a certain product, and the category to which the product belongs can be determined based on the advertisement content (e.g., image, video, text, etc.). For example, the product category can be a daily necessities category, a beauty product category, a furniture category, an electronic product category, etc. Note that only some example categories are given here, and in actual scenarios, a category library can be configured according to various standards, according to different granularities, and the target category to which the current target recommendation content belongs can be determined from the category library. In some embodiments, the category library can be configured as a category hierarchy, which includes categories having a plurality of hierarchical relationships. In this way, an object corresponding to the recommendation content can be classified into a large category (also referred to as a parent category) corresponding to a higher level, and can also be classified into a specific subcategory under the parent category. For example, if a product is a face cream, the product can be classified into a beauty product category (parent category) in a higher category hierarchy, and can also be more specifically classified into a skin care subcategory under the beauty product category. It can be understood that the hierarchical construction of different objects can depend on specific scenarios, and embodiments of the present disclosure do not limit this aspect. In general, when dividing the target category corresponding to the target recommendation content, the target category can be a specific category in the category hierarchy, or can include categories corresponding to a plurality of levels in the target hierarchy. In some embodiments, in addition to the target category to which the target recommendation content itself belongs, one or more categories related to the target category can also be concerned. This is because, as will be discussed below, in embodiments of the present disclosure, it is desired to obtain historical behavior data in the corresponding category according to the category information of the target recommendation content, in order to help determine the target audience group for the target recommendation content. In some embodiments, the historical behavior data of the related categories can also affect the determination of the audience group for the recommendation content of the target category. In determining the one or more categories related to the target category, the respective degrees of association of the target category and a plurality of candidate categories can be determined based on the degree of association between the historical behavior data of the historical recommendation content of the target category when provided to the audience group and the historical behavior data of the historical recommendation content of the plurality of candidate categories when provided to the audience group. The degree of association here may, for example, be based on the time association of the behavior corresponding to the recommendation content of different categories.Then, based on the degree of association of the target category with each of the plurality of candidate categories, at least one category that is relevant to the target category is determined from the plurality of candidate categories. For example, if the user's purchase behavior of the diaper category recommendation content is detected, and at the same time or within a short period of time, the user's purchase behavior of the beer category recommendation content is detected, it can be determined that the degree of association between the diaper category and the beer category is high. That is, the diaper category and the beer category are relevant categories to each other. Of course, in addition to the categories exemplified here, more relevant categories can be found by analyzing historical behavior data. In block 230, the recommendation management system 150 obtains historical behavior data associated with the category indicated by the category information, the historical behavior data indicating at least one behavior performed by the audience group on the recommendation content of the corresponding category when the recommendation content is provided to the audience group. As mentioned earlier, after the recommendation content is provided to the audience group, one or more behaviors of the audience group on the recommendation content can be obtained, including but not limited to: clicking, searching, liking, clicking interest, commenting, buying, downloading, etc. In some embodiments, the historical behavior data indicates a plurality of behaviors performed by the audience group on the recommendation content when the recommendation content of the corresponding category is provided to the audience group. Because the target recommendation content has not been provided or has not been provided for a long time to the audience group, the behavior of the audience group related to the target recommendation content cannot be obtained. The historical behavior data associated with a certain category can indicate the degree of interest of the audience group on the recommendation content of the category, the possibility of performing a certain behavior, etc. Therefore, in embodiments of the present disclosure, by referring to the historical behavior data corresponding to other recommendation contents under the same category or relevant category, a more accurate and more matched target audience group for the target recommendation content can be determined. In some embodiments, the obtained historical behavior data is collected within a predetermined time period, for example, historical behavior data collected within a period of time before the time of receiving the recommendation request. With updated historical behavior data, the conversion tendency of the audience group for the recommendation content at the current time can be captured more timely and accurately. In some embodiments, one or more behaviors of interest in the delivery of the target recommendation content can be determined, and historical behavior data corresponding to the one or more behaviors of interest can be collected. In some embodiments, when obtaining the historical behavior data, in addition to the category information of the target recommendation content, the content delivery platform where the target recommendation content is to be provided, the geographic area where the target recommendation content is to be provided, etc. can also be considered. For example, the recommendation contents of various categories (these recommendation contents are referred to as similar recommendation contents or relevant recommendation contents of the target recommendation content) can be determined from the same content delivery platform, and / or the same delivery geographic area, and the historical behavior data corresponding to the determined recommendation contents can be obtained.At block 240, the recommendation management system 150 determines a target audience group corresponding to the target recommendation content based on historical behavior data, for providing the target recommendation content to at least part of the target audience group. The historical behavior data can be used to characterize various conversion behaviors performed by the audience groups of the recommendation content corresponding to the target category and / or its related categories. With the historical behavior data, by referring to the features of the audience groups performing certain behaviors during the historical delivery of the recommendation content of each category, it can be better determined how to determine the target audience group for the target recommendation content. In some embodiments, the recommendation management system 150 can determine candidate audience groups corresponding to the historical behavior data associated with each category indicated by the category information, and determine the target audience group based on the determined candidate audience groups. This is because each category (the target category or its related categories) can help determine the audience groups that have performed certain behaviors on the recommendation content, and then by analyzing these audience groups, the target audience group corresponding to the target recommendation content can be determined. Since the candidate audience groups are determined from the audience groups of the recommendation content under the same or related categories as the target recommendation content, and these audience groups have all performed certain behavior(s) on their respective recommendation content, it means that these audience groups are satisfied with the recommendation of the recommendation content of the same or related categories, and are likely to perform conversion. In some embodiments, the recommendation management system 150 can determine target label information O corresponding to the target audience group based on the label information corresponding to the determined candidate audience groups. In the recommendation system, different audience groups can be labeled with the characteristics or features of the audience group by label information. Audience groups usually with the same label information can also have the same conversion of the recommendation content. Therefore, the label information corresponding to the candidate audience groups screened by the category information can reasonably determine what kind of label information the target audience group with which the target recommendation content can be provided. With the target label information, the target audience group to be used to deliver the target recommendation content can be identified among other audience groups. In some embodiments, the recommendation management system 150 can simply screen the target audience group from the determined candidate audience groups. The target audience group can be part or all of the candidate audience groups. In some embodiments, the recommendation management system 150 can determine the number of audiences of the target audience group based on the number of audiences of the determined candidate audience groups. When providing the target recommendation content, the recommendation requester can not be able to determine the scale of the provision of the target recommendation content. With the candidate audience groups determined by the category information, the scale of the target audience group can also be reasonably determined in order to optimize the delivery strategy of the target recommendation content.If the target category or the related categories indicated by the category information are categorized in a category hierarchy, the audience group can also be expanded or narrowed in the corresponding categories at a higher or lower category hierarchy. In some embodiments, if the number of the candidate audience groups determined based on the category information of the target recommendation content is too large (e.g., exceeds a threshold number), the recommendation management system 150 can determine a subcategory of the target category in the category hierarchy. Then, the candidate audience groups associated with the subcategory are determined from the candidate audience groups corresponding to each category indicated by the category information, so as to achieve further narrowing of the candidate audience groups, because the subcategory is usually associated with fewer candidate audience groups. Then, the recommendation management system 150 can determine the target audience group from the candidate audience groups associated with the subcategory, for example, can directly filter the target audience group therefrom, or determine the label information corresponding to the target audience group, or determine the number of the target audience group. In some embodiments, if the number of the candidate audience groups determined based on the category information of the target recommendation content is too small (e.g., is lower than a threshold number), the recommendation management system 150 can determine a parent category including the target category in the category hierarchy. This can expand the category range of the target recommendation content. Then the recommendation management system 150 can determine the candidate audience groups associated with the parent category. Specifically, the recommendation management system 150 can obtain the historical behavior data associated with the parent category, and determine the candidate audience groups associated with the parent category from the historical behavior data. This can achieve further expansion of the candidate audience groups, because the parent category is usually associated with fewer candidate audience groups. The recommendation management system 150 can determine the target audience group from the candidate audience groups associated with the parent category. In some embodiments, if the historical behavior data indicates a plurality of behaviors performed by the audience group on the recommendation content when the recommendation content of the corresponding category is provided to the audience group, each corresponding weight can also be assigned to the plurality of behaviors. In determining the target audience group, the target audience group can be determined from the candidate audience groups corresponding to each category based on the respective weights of the plurality of behaviors. In some examples, for the behaviors with higher weights, more target audience groups can be determined from the candidate audience groups performing the behaviors, or the label information corresponding to the candidate audience groups is considered more. For the behaviors with lower weights, the target audience groups can not be considered or are considered less from the candidate audience groups performing the behaviors, or the label information corresponding to the candidate audience groups is considered less or not.In some embodiments, the target audience group can be determined from candidate audience groups that more frequently perform a particular behavior (e.g., a search behavior, a purchase behavior) based on historical behavior data, as these audience groups show a higher conversion likelihood for a particular category of recommendation content. Alternatively or additionally, if the category information of the target recommendation content indicates both a target category and at least one category related to the target category, then similarly, a corresponding weight can be assigned to the target category and the related categories, respectively. The target category corresponds to a first weight and the at least one category related to the target category corresponds to a second weight, where the first weight is higher than the second weight. This is because the target category indicates the category to which the target recommendation content itself belongs. In some embodiments, if there are multiple related categories, then different second weights can be configured for the multiple related categories according to the degree of relevance between the related categories and the target category. In determining the target audience group, the target audience group can be determined from the candidate audience group corresponding to each category based on the weight corresponding to each category. In some examples, for a category with a higher weight, more target audience groups can be determined from the candidate audience group associated with the category or more consideration can be given to the label information corresponding to the candidate audience group. For a category with a lower weight, less or no consideration can be given to the candidate audience group associated with the category in determining the target audience group or less or no consideration can be given to the label information corresponding to the candidate audience group. The above embodiments discuss the use of historical behavior data in determining the target audience group for a current target recommendation content. These embodiments can be implemented individually or in combination to determine a target audience group that is more matched to the target recommendation content and is more likely to perform a conversion. FIG. 3 illustrates a flowchart of a method 300 for content recommendation according to some embodiments of the present disclosure. The method 300 can be implemented by the recommendation management system 150 shown in FIG. 1, for example. At block 310, the recommendation management system 150 determines, in response to receiving a provision request for a target recommendation content, category information corresponding to the target recommendation content, the category information indicating a target category to which the target recommendation content is classified and / or at least one category related to the target category. At block 320, the recommendation management system 150 obtains historical behavior data associated with the categories indicated by the category information, the historical behavior data indicating at least one behavior performed by an audience group on a recommendation content of a corresponding category when the recommendation content of the corresponding category is provided to the audience group. At block 330, the recommendation management system 150 determines, based on the historical behavior data, a target audience group corresponding to the target recommendation content for providing the target recommendation content to at least part of the target audience group.In some embodiments, determining the category information corresponding to the target recommended content comprises: determining an association degree between historical behavior data of the target category's historical recommended content when provided to the audience group and historical behavior data of the plurality of candidate categories' historical recommended content when provided to the audience group; and determining the target category's association degree with each of the plurality of candidate categories based on the association degree, and determining at least one category related to the target category from the plurality of candidate categories based on the association degree of the target category with each of the plurality of candidate categories. In some embodiments, determining the target audience group corresponding to the target recommended content based on the determined candidate audience groups comprises: determining a candidate audience group corresponding to each category indicated by the category information based on the historical behavior data; and determining the target audience group based on the determined candidate audience groups. In some embodiments, determining the target audience group based on the determined candidate audience groups comprises performing at least one of: determining target label information corresponding to the target audience group based on label information corresponding to the determined candidate audience groups; determining a number of audiences of the target audience group based on a number of audiences of the determined candidate audience groups; and filtering the target audience group from the determined candidate audience groups. In some embodiments, determining the target audience group based on the determined candidate audience groups comprises: in response to the number of audiences of the determined candidate audience groups exceeding a threshold number, determining a subcategory of the target category at a category level; determining a candidate audience group associated with the subcategory from the candidate audience groups corresponding to each category indicated by the category information; and determining the target audience group from the candidate audience groups associated with the subcategory. In some embodiments, determining the target audience group based on the determined candidate audience groups comprises: in response to the number of audiences of the determined candidate audience groups being less than a threshold number, determining a parent category including the target category at a category level; determining a candidate audience group associated with the parent category; and determining the target audience group from the candidate audience group associated with the parent category. In some embodiments, the historical behavior data indicates a plurality of behaviors performed by the audience group on the recommended content of the corresponding category when the recommended content is provided to the audience group. In some embodiments, determining the target audience group based on the determined candidate audience groups comprises: determining the target audience group from the candidate audience groups corresponding to each category based on a weight corresponding to each of the plurality of behaviors. In some embodiments, the category information indicates the target category and at least one category related to the target category. In some embodiments, determining the target audience group based on the determined candidate audience groups comprises: determining the target audience group from the candidate audience groups corresponding to each category based on a first weight corresponding to the target category and a second weight corresponding to the at least one category related to the target category, wherein the first weight is higher than the second weight.In some embodiments, obtaining the historical behavior data associated with the category indicated by the category information comprises: obtaining the historical behavior data associated with the category indicated by the category information collected in a predetermined time period. FIG. 4 shows a schematic structural block diagram of an apparatus 400 for content recommendation according to some embodiments of the present disclosure. The apparatus 400 can be implemented as or included in the recommendation management system 150. Various modules / components in the apparatus 400 can be implemented by hardware, software, firmware, or any combination thereof. As shown, the apparatus 400 includes a category determining module 410 configured to, in response to receiving a provision request for target recommendation content, determine category information corresponding to the target recommendation content, the category information indicating a target category to which the target recommendation content is classified, and / or at least one category related to the target category. The apparatus 400 further includes a data obtaining module 420 configured to obtain historical behavior data associated with the category indicated by the category information, the historical behavior data indicating at least one behavior performed by an audience group on a recommendation content of the corresponding category when the recommendation content is provided to the audience group. The apparatus 400 further includes an audience determining module 430 configured to determine, based on the historical behavior data, a target audience group corresponding to the target recommendation content for providing the target recommendation content to at least part of the target audience group. In some embodiments, the category determining module 410 is further configured to: determine respective degrees of association of the target category with a plurality of candidate categories based on a degree of association between the historical behavior data of the historical recommendation content of the target category when provided to the audience group and the historical behavior data of the historical recommendation content of the plurality of candidate categories when provided to the audience group; and determine the at least one category related to the target category from the plurality of candidate categories based on the respective degrees of association of the target category with the plurality of candidate categories. In some embodiments, the audience determining module 430 is further configured to include: determining a candidate audience group corresponding to the historical behavior data associated with each category indicated by the category information; and determining the target audience group based on the determined candidate audience groups. In some embodiments, the audience determining module 430 is further configured to: determine target label information corresponding to the target audience group based on label information corresponding to the determined candidate audience groups; determine a number of audiences of the target audience group based on a number of audiences of the determined candidate audience groups; and filter the target audience group from the determined candidate audience groups.In some embodiments, the audience determination module 430 is further configured to: in response to the determined number of audiences of the candidate audience group exceeding a threshold number, determine a subcategory of the target recommendation content under a category in a category hierarchy; determine, from the candidate audience group corresponding to each category indicated by the category information, a candidate audience group associated with the subcategory; and determine the target audience group from the candidate audience group associated with the subcategory. In some embodiments, the audience determination module 430 is further configured to: in response to the determined number of audiences of the candidate audience group being lower than the threshold number, determine a parent category including the category under which the target category is located in the category hierarchy; determine a candidate audience group associated with the parent category; and determine the target audience group from the candidate audience group associated with the parent category. In some embodiments, the historical behavior data indicates a plurality of behaviors performed by the audience group on the recommendation content of the corresponding category when the recommendation content is provided to the audience group. In some embodiments, the audience determination module 430 is further configured to: determine the target audience group from the candidate audience group corresponding to each category based on a weight corresponding to each of the plurality of behaviors. In some embodiments, the category information indicates the target category and at least one category related to the target category. In some embodiments, the audience determination module 430 is further configured to: determine the target audience group from the candidate audience group corresponding to each category based on a first weight corresponding to the target category and a second weight corresponding to the at least one category related to the target category, wherein the first weight is higher than the second weight. In some embodiments, the data obtaining module 420 is further configured to: obtain the historical behavior data associated with the categories indicated by the category information collected within a predetermined time period. FIG. 5 illustrates a block diagram of an electronic device 500 in which one or more embodiments of the present disclosure can be implemented. It should be understood that the electronic device 500 illustrated in FIG. 5 is merely exemplary and should not be construed as limiting the functionality and scope of the embodiments described herein. The electronic device 500 illustrated in FIG. 5 can be used to implement the client device 130, or the content delivery platform 110 and / or the recommendation management system 150 (or individual components therein). The electronic device 500 can include or be implemented as the apparatus 400 of FIG. 4. As shown in FIG. 5, the electronic device 500 is in the form of a general computing device. The components of the electronic device 500 can include, but are not limited to, one or more processors or processing units 510, a memory 520, a storage device 530, one or more communication units 540, one or more input devices 550, and one or more output devices 560. The processing unit 510 can be a real or virtual processor and is capable of performing various processes according to programs stored in the memory 520.In a multi-processor system, multiple processing units execute computer-executable instructions in parallel to improve the parallel processing capabilities of electronic device 500. Electronic device 500 typically includes multiple computer storage media. Such media can be any available media that is accessible by electronic device 500 and includes both volatile and non-volatile media, removable and non-removable media. Memory 520 can be volatile memory (e.g., registers, cache, random access memory (RAM)), non-volatile memory (e.g., read only memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory), or some combination thereof. Storage 530 can be a removable or non-removable media, and can include machine readable media such as a flash drive, a magnetic disk drive, or any other media that can be used to store information and / or data (e.g., training data for training) and that can be accessed by electronic device 500. Electronic device 500 can further include additional removable / non-removable, volatile / non-volatile storage media. Although not shown in FIG. 5, a disk drive for reading from or writing to a removable, non-volatile magnetic disk (e.g., a “floppy disk”), and an optical disk drive for reading from or writing to a removable, non-volatile optical disk (e.g., a CD-ROM) can be provided. In such instances, each drive can be connected to the bus (not shown) by one or more data media interfaces. Memory 520 can include computer program product 525 having one or more program modules configured to carry out the various methods or actions of the various embodiments of the present disclosure. Communication unit 540 enables communications with other electronic devices over a communication medium. Additionally, the functionality of the components of electronic device 500 can be implemented in a single computing cluster or a plurality of computer machines that are capable of communicating over a communication connection. Thus, electronic device 500 can operate in a networked environment using logical connections to one or more other servers, network personal computers (PCs), or another network nodes in the networking environment. Input device 550 can be one or more input devices, such as a mouse, a keyboard, a trackball, etc. Output device 560 can be one or more output devices, such as a display, a speaker, a printer, etc. Electronic device 500 can also communicate with one or more external devices (not shown) such as a storage device, a display device, etc. through communication unit 540, one or more devices that enable a user to interact with electronic device 500, or any devices (e.g., network cards, modems, etc.) that enable electronic device 500 to communicate with one or more other electronic devices.Such communication can be carried out via an input / output (I / O) interface (not shown). According to an exemplary implementation of the present disclosure, a computer readable storage medium is provided having computer executable instructions stored thereon, where the computer executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is also provided, tangibly stored on a non-transitory computer readable medium and comprising computer executable instructions, where the computer executable instructions are executed by a processor to implement the method described above. Various aspects of the disclosure can be described in the context of flow diagrams and / or block diagrams that illustrate the functions and operations of methods, apparatuses, devices, and computer program products according to implementations of the present disclosure. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer readable program instructions. These computer readable program instructions can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flow diagrams and / or block diagrams. These computer readable program instructions can also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and / or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including a manufacture including instructions which implement aspects of the function / act specified in the flow diagrams and / or block diagrams. The computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer, other programmable data processing apparatus, or other device implement the functions / acts specified in the flow diagrams and / or block diagrams. The flow diagrams and block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various implementations of the present disclosure. In this regard, each block in the flow diagrams and block diagrams can represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical functions ("instructions"). In some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession can in fact be executed substantially concurrently or the blocks can sometimes be executed in the reverse order, depending upon the functionality involved.It is also important to note that each block of the flowchart and / or block diagram illustrations, and combinations of blocks in the flowchart and / or block diagram illustrations, can be implemented by dedicated-function hardware-based systems that perform the specified functions or acts, or combinations of dedicated-function hardware and computer instructions. Having thus described the functionality of various implementations of the disclosure, it is to be appreciated that the above description has been presented for purposes of example and that various modifications and variations can be made to the implementations disclosed without departing from the scope of the description. Many modifications and variations are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the disclosure various implementations can be practiced otherwise than as specifically outlined above. The scope of the disclosure is defined by the appended claims, rather than the preceding description, and all changes that come within the meaning and range of equivalents of the claims are to be embraced within their scope.

Claims

CLAIM 1. A method for content recommendation, comprising: in response to receiving the request for provision of the target recommended content, determining category information corresponding to the target recommended content, the category information indicating a target category to which the target recommended content is classified, and / or at least one category related to the target category; obtaining historical behavior data associated with the categories indicated by the category information, the historical behavior data indicating at least one behavior performed by an audience group on a recommended content of a corresponding category when the recommended content of the corresponding category is provided to the audience group; and based on the historical behavior data, determining a target audience group corresponding to the target recommended content for providing the target recommended content to at least part of the target audience group.

2. The method of claim 1, wherein determining the category information corresponding to the target recommended content comprises: determining, based on a degree of association between the historical behavior data of the historical recommended content of the target category when provided to an audience group and the historical behavior data of the historical recommended content of each of a plurality of candidate categories when provided to an audience group, a degree of association of the target category with each of the plurality of candidate categories; and based on the degree of association of the target category with each of the plurality of candidate categories, determining at least one category related to the target category from the plurality of candidate categories.

3. The method of claim 1, wherein determining, based on the historical behavior data, a target audience group to which the target recommended content corresponds comprises: determining a candidate audience group corresponding to the historical behavior data associated with each category indicated by the category information; and determining the target audience group based on the determined candidate audience groups.

4. The method of claim 3, wherein determining the target audience group based on the determined candidate audience groups comprises performing at least one of: determining target tag information corresponding to the target audience group based on tag information corresponding to the determined candidate audience groups; determining a number of audiences of the target audience group based on a number of audiences of the determined candidate audience groups; and filtering the target audience group from the determined candidate audience groups.

5. The method of claim 3, wherein determining the target audience group based on the determined candidate audience group comprises: in response to the number of audiences of the determined candidate audience groups exceeding a threshold number, determining a subcategory of the target category at a category level; determining a candidate audience group associated with the subcategory from the candidate audience groups corresponding to each category indicated by the category information; and determining the target audience group from the candidate audience group associated with the subcategory.

6. The method of claim 3, wherein determining the target audience group based on the determined candidate audience groups comprises: in response to the number of audiences of the determined candidate audience groups being lower than a threshold number, determining a parent category of the target category at a category level; determining a candidate audience group associated with the parent category; and determining the target audience group from the candidate audience group associated with the parent category.

7. The method of claim 3, wherein the historical behavior data indicates a plurality of behaviors performed by an audience group when a respective category of recommended content is provided to the audience group, and wherein determining the target audience group based on the determined candidate audience group comprises: determining the target audience group from the candidate audience groups corresponding to each category based on weights corresponding to each of the plurality of behaviors.

8. The method of claim 3, wherein the category information indicates the target category and at least one category related to the target category, and wherein determining the target audience group based on the determined candidate audience groups comprises: determining the target audience group from the candidate audience groups corresponding to each category based on a first weight corresponding to the target category and a second weight corresponding to at least one category related to the target category, wherein the first weight is higher than the second weight.

9. The method of claim 1, wherein obtaining historical behavior data associated with a category indicated by the category information comprises: obtaining historical behavior data associated with the category indicated by the category information, the historical behavior data indicating at least one behavior performed by a target audience group on a recommendation content of the corresponding category when the recommendation content is provided to the target audience group; 10. An apparatus for content recommendation, comprising: a category determining module configured to determine, in response to receiving a provision request for a target recommendation content, category information corresponding to the target recommendation content, the category information indicating a target category to which the target recommendation content is classified and / or at least one category related to the target category; a data obtaining module configured to obtain historical behavior data associated with the category indicated by the category information, the historical behavior data indicating at least one behavior performed by an audience group on a recommendation content of the corresponding category when the recommendation content is provided to the audience group; and an audience determining module configured to determine, based on the historical behavior data, a target audience group corresponding to the target recommendation content, for providing the target recommendation content to at least part of the target audience group.

11. An electronic device, comprising: at least one processing unit; and at least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, cause the device to perform the method according to any one of claims 1 to 9.

12. A computer-readable storage medium having computer-executable instructions stored thereon, the computer-executable instructions, when executed by a processor, implementing the method according to any one of claims 1 to 9.

13. A computer program product tangibly stored in a computer storage medium and comprising computer-executable instructions that, when executed by a device, cause the device to perform the method according to any one of claims 1 to 9. ​

Images