Video recommendation method and device, electronic equipment and readable storage medium

Abstract

The invention relates to a video recommendation method and device, electronic equipment and a readable storage medium, and belongs to the technical field of computers. The method comprises the following steps: acquiring user feature data of a target user and video feature data of each first video; the first video is a to-be-recommended video corresponding to the first-level page; inputting the user feature data and the video feature data of each first video into a pre-trained video recommendation model, and outputting a user behavior score and a secondary page view score corresponding to each first video; wherein the score of the second-level page view of each first video represents the predicted number of second videos continuously browsed in a second-level page after the target user jumps to the second-level page based on the first video; the second video is a recommended video corresponding to the second-level page; and determining a recommended video corresponding to the first-level page in each first video according to the user behavior score and the second-level page view score corresponding to each first video. By adopting the method and the device, the video page view in the second-level page can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a video recommendation method, device, electronic device and readable storage medium. Background Art

[0002] At present, the architecture of short video platforms usually includes a first-level page and a second-level page. The first-level page is the homepage of the short video platform. The short video platform displays video content that can quickly attract users' attention on the first-level page. The second-level page is the video details page, which is entered by jumping from the video link on the first-level page. The short video platform displays the complete video content on the second-level page, and further displays other video content based on user operations (such as sliding down).

[0003] However, since the video recommendation model used by short video platforms on the first-level pages mainly focuses on quickly attracting user attention, users may find that the video content does not meet their expectations after clicking on the video content and quickly exit the video content, which in turn leads to lower video views on the second-level pages. Summary of the invention

[0004] Based on this, it is necessary to provide a video recommendation method, device, electronic device and readable storage medium to address the above technical issues.

[0005] In a first aspect, a video recommendation method is provided, the method comprising:

[0006] Obtaining user feature data of a target user and video feature data of each first video; wherein the first video is a video to be recommended corresponding to a first-level page;

[0007] The user feature data and the video feature data of each of the first videos are input into a pre-trained video recommendation model, and a user behavior score and a secondary page view score corresponding to each of the first videos are output; wherein the secondary page view score of each of the first videos represents the predicted number of the target user who continues to browse the second video on the secondary page after jumping to the secondary page based on the first video; the second video is a recommended video corresponding to the secondary page;

[0008] According to the user behavior score and the secondary page view score corresponding to each of the first videos, a recommended video corresponding to the primary page is determined in each of the first videos.

[0009] As an optional implementation manner, before inputting the user feature data and the video feature data of each of the first videos into a pre-trained video recommendation model, the method further includes:

[0010] Acquire first characteristic data of the target user; wherein the first characteristic data is characteristic data obtained through function interaction;

[0011] The step of inputting the user feature data and the video feature data of each of the first videos into a pre-trained video recommendation model includes:

[0012] The first feature data, the user feature data, and the video feature data of each of the first videos are input into a pre-trained video recommendation model.

[0013] As an optional implementation manner, determining the recommended video corresponding to the first-level page in each of the first videos according to the user behavior score and the second-level page view score corresponding to each of the first videos includes:

[0014] Determine a video recommendation score for each of the first videos according to the user behavior score and the secondary page view score corresponding to each of the first videos;

[0015] Among the first videos, a preset number of the first videos are selected in descending order of video recommendation scores and determined as recommended videos corresponding to the first-level page.

[0016] As an optional implementation, the method further includes:

[0017] Acquire training sample data and label data; wherein the training sample data includes user feature data of a sample user and video feature data of a third video browsed by the sample user in the primary page, and the label data includes user behavior data corresponding to the third video and the number of views of a fourth video, wherein the third video is a video in the primary page browsed by the sample user; and the fourth video is a video that the sample user continues to browse in the secondary page after jumping to the secondary page based on the third video;

[0018] The initial video recommendation model is trained according to the training sample data and the label data to obtain a trained video recommendation model.

[0019] As an optional implementation, the training sample data also includes second feature data; wherein the second feature data is feature data obtained through functional interaction when the sample user browses the third video in the first-level page.

[0020] As an optional implementation manner, the obtaining of training sample data and label data includes:

[0021] In response to a trigger operation of the sample user on the third video in the first-level page, obtaining user feature data of the sample user, the second feature data, and video feature data and user behavior data of the third video;

[0022] According to the user ID of the sample user and the video ID of the third video, searching the video browsing log of the sample user for a candidate video that belongs to the secondary page and is associated with the user ID and the video ID;

[0023] Based on the queried candidate videos, the number of views of the fourth video is determined.

[0024] As an optional implementation manner, determining the number of views of the fourth video based on the queried candidate videos includes:

[0025] Determine the number of candidate videos found as the number of views of the fourth video; or,

[0026] Among the queried candidate videos, the number of candidate videos whose browsing time is greater than a preset browsing time threshold is determined as the browsing volume of the fourth video.

[0027] In a second aspect, a video recommendation device is provided, the device comprising:

[0028] A first acquisition module is used to acquire user feature data of a target user and video feature data of each first video; wherein the first video is a video to be recommended corresponding to a first-level page;

[0029] An output module is used to input the user feature data and the video feature data of each of the first videos into a pre-trained video recommendation model, and output a user behavior score and a secondary page view score corresponding to each of the first videos; wherein the secondary page view score of each of the first videos represents the predicted number of the target user who continues to browse the second video in the secondary page after jumping to the secondary page based on the first video; the second video is a recommended video corresponding to the secondary page;

[0030] A determination module is used to determine the recommended video corresponding to the first-level page in each of the first videos according to the user behavior score and the second-level page view score corresponding to each of the first videos.

[0031] As an optional implementation, the device further includes:

[0032] A second acquisition module is used to acquire first feature data of the target user; wherein the first feature data is a feature obtained through functional interaction;

[0033] The output module is specifically used for:

[0034] The first feature data, the user feature data, and the video feature data of each of the first videos are input into a pre-trained video recommendation model.

[0035] As an optional implementation manner, the determining module is specifically configured to:

[0036] Determine a video recommendation score for each of the first videos according to the user behavior score and the secondary page view score corresponding to each of the first videos;

[0037] Among the first videos, a preset number of the first videos are selected in descending order of video recommendation scores and determined as recommended videos corresponding to the first-level page.

[0038] As an optional implementation, the device further includes:

[0039] A third acquisition module is used to acquire training sample data and label data; wherein the training sample data includes user feature data of a sample user and video feature data of a third video browsed by the sample user in the first-level page, and the label data includes user behavior data corresponding to the third video and the number of views of a fourth video, wherein the third video is a video in the first-level page browsed by the sample user; and the fourth video is a video that the sample user continues to browse in the second-level page after jumping to the second-level page based on the third video;

[0040] The training module is used to train the initial video recommendation model according to the training sample data and the label data to obtain a trained video recommendation model.

[0041] As an optional implementation, the training sample data also includes second feature data; wherein the second feature data is feature data obtained through functional interaction when the sample user browses the third video in the first-level page.

[0042] As an optional implementation manner, the acquisition module is specifically used to:

[0043] In response to a trigger operation of the sample user on the third video in the first-level page, obtaining user feature data of the sample user, the second feature data, and video feature data and user behavior data of the third video;

[0044] According to the user ID of the sample user and the video ID of the third video, searching the video browsing log of the sample user for a candidate video that belongs to the secondary page and is associated with the user ID and the video ID;

[0045] Based on the queried candidate videos, the number of views of the fourth video is determined.

[0046] As an optional implementation manner, the acquisition module is specifically used to:

[0047] Determine the number of candidate videos found as the number of views of the fourth video; or,

[0048] Among the queried candidate videos, the number of candidate videos whose browsing time is greater than a preset browsing time threshold is determined as the browsing volume of the fourth video.

[0049] According to a third aspect, an electronic device is provided, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; the memory is used to store computer programs; and the processor is used to implement any method described in the first aspect when executing the program stored in the memory.

[0050] According to a fourth aspect, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the method described in any one of the first aspects is implemented.

[0051] The present application provides a video recommendation method, device, electronic device, and readable storage medium. The technical solution provided by the embodiments of the present application brings at least the following beneficial effects:

[0052] When training the video recommendation model, this scheme adds the number of videos that the sample user continues to browse on the secondary page after jumping to the secondary page based on the third video, so that the video recommendation model gives positive incentives to the primary page videos that are consumed more on the secondary page, thereby reducing the recommendation of videos with low relevance to the secondary page.

[0053] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 A flowchart of a video recommendation method provided in an embodiment of the present application;

[0056] Figure 2 A schematic diagram of a network structure of a video recommendation model provided in an embodiment of the present application;

[0057] Figure 3 A flowchart of a method for training a video recommendation model provided in an embodiment of the present application;

[0058] Figure 4 A schematic diagram of a video recommendation model training provided in an embodiment of the present application;

[0059] Figure 5 A schematic diagram of the structure of a video recommendation device provided in an embodiment of the present application;

[0060] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0062] At present, the architecture of short video platforms usually includes a first-level page and a second-level page. The first-level page is the homepage of the short video platform. The short video platform displays video content that can quickly attract users' attention on the first-level page. The second-level page is the video details page, which is entered by jumping from the video link on the first-level page. The short video platform displays the complete video content on the second-level page, and further displays other video content based on user operations (such as sliding down).

[0063] However, since the video recommendation model used by short video platforms on the first-level pages mainly focuses on quickly attracting user attention, users may find that the video content does not meet their expectations after clicking on the video content and quickly exit the video content, which in turn leads to lower video views on the second-level pages.

[0064] The present application provides a video recommendation method. The following will describe in detail a video recommendation method provided by an embodiment of the present application in combination with a specific implementation method. Figure 1 This is a flowchart of a video recommendation method provided in an embodiment of the present application. This embodiment uses the method applied to a computer device as an example. The computer device may be a terminal or a server. It is understandable that the method may also be applied to a system including a terminal and a server and implemented through the interaction between the terminal and the server. Figure 1 As shown, the specific steps are as follows:

[0065] Step 101: Obtain user feature data of a target user and video feature data of each first video.

[0066] Among them, the first video is the video to be recommended corresponding to the first-level page.

[0067] In implementation, the short video recommendation system usually includes multiple levels of pages, such as a first-level page and a second-level page. Among them, the first-level page can include multiple display areas, such as a homepage display area, a double-column display area, etc., and different display areas can be used to display videos with different recommendation levels. When a user (which can be called a target user) wants to access the first-level page, the target user can execute the trigger instruction corresponding to the first-level page (such as opening the client, or clicking the access entrance corresponding to the first-level page, etc.) to enter the first-level page to browse the video. In order to improve the accuracy of video recommendation, the computer device can obtain the user feature data of the target user and the video feature data of each first video to be recommended when the preset conditions are met. Among them, the preset conditions can be receiving the trigger instruction corresponding to the first-level page from the target user, or reaching a preset period. The user feature data may include one or more of the user identification, user age, user gender, member attributes (such as whether to be a member, member level, etc.), login attributes (such as whether to log in, the type of login device, etc.), browsing history behavior, click history behavior and user long-term and short-term portraits. The video feature data includes one or more of the video identification, video attributes, video creator identification and video creator main attributes.

[0068] Step 102: input the user feature data and the video feature data of each first video into a pre-trained video recommendation model, and output the user behavior score and the secondary page view score corresponding to each first video.

[0069] Among them, the secondary page view score of each first video represents the predicted number of target users who continue to browse the second video in the secondary page after jumping to the secondary page based on the first video; the second video is the recommended video corresponding to the secondary page.

[0070] In implementation, after the computer device obtains the user feature data of the target user and the video feature data of each first video, the user feature data and the video feature data of each first video can be input into a pre-trained video recommendation model to obtain a user behavior score and a secondary page view score corresponding to each first video. Optionally, the user behavior score is used to characterize the predicted probability of the user performing various operation behaviors on the first video, and the operation behavior can include multiple dimensions of behavior that are used to reflect the depth of interaction between the user and the first video. For example, the operation behavior may include but is not limited to one or more of clicking on the video, browsing for a certain length of time, watching to completion, liking, commenting, collecting, and forwarding. Accordingly, the user behavior score may include but is not limited to one or more of the user's predicted click probability, predicted viewing time, completion probability, liking probability, comment probability, collection probability, and forwarding probability for the first video, which is not limited in the embodiments of the present application.

[0071] Optionally, the video recommendation model may use a DNN (Deep Neural Networks) model, or other types of network models, which are not limited in the present embodiment. Figure 2 , which is a schematic diagram of a network structure of a video recommendation model provided in an embodiment of the present application. The network includes an embedding layer and multiple ReLU (Rectified Linear Unit, rectified linear units).

[0072] Step 103, determining a recommended video corresponding to the primary page in each first video according to the user behavior score and the secondary page view score corresponding to each first video.

[0073] In implementation, the computer device can determine the first video that meets the screening condition among the first videos according to the user behavior score and the secondary page view score corresponding to each first video, and use the determined first video as the recommended video corresponding to the primary page. In this way, the client can display the recommended video on the primary page so that the user can browse the recommended video.

[0074] In an embodiment of the present application, a video recommendation model is used to predict the user behavior score and the secondary page view score corresponding to each first video, and then the recommended video corresponding to the primary page is determined in each first video in combination with the user behavior score and the secondary page view score, thereby improving the matching degree between the recommended video and the user's preference, and the probability of the user consuming the secondary page based on the recommended video, thereby increasing the video views in the secondary page.

[0075] Moreover, in this solution, by adding the secondary page consumption depth fitting target (i.e., the secondary page view score), the subsequent consumption depth of the declining users can be improved. This method effectively improves the effect of video recommendation with a smaller parameter amount, occupies less resources, has higher processing efficiency and higher real-time performance. Moreover, it can enrich the core consumption indicators of the first and second level pages, give play to the advantages of multi-terminal big data integration, effectively improve the core consumption indicators of multi-terminal online, and improve the user experience.

[0076] In some embodiments, the short video recommendation system generally includes multiple levels of pages, for example, at least a first-level page, a second-level page, and a third-level page.

[0077] A first-level page may include multiple display areas, such as a homepage display area, a double-column display area, etc. Different display areas may be used to display videos with different recommendation levels. When a user (referred to as a target user) wants to access a first-level page, the target user may execute a trigger instruction corresponding to the first-level page (such as opening a client, or clicking on an access entry corresponding to the first-level page, etc.) to enter the first-level page to browse videos.

[0078] The secondary page is the video details page, which is accessed by jumping to the video link in the primary page, and further displays other video content based on user operations (such as swiping down).

[0079] The third-level page is the page that jumps to after being triggered by other video content in the second-level page (for example, the user clicks on other video content in the second-level page). The third-level page displays the video details of other video content clicked by the user in the second-level page, and further displays more video content based on the user's operation (such as sliding down).

[0080] In some embodiments, in step 102, before inputting the user feature data and the video feature data of each first video into the pre-trained video recommendation model, the method further includes: obtaining the first feature data of the target user. Accordingly, inputting the user feature data and the video feature data of each first video into the pre-trained video recommendation model includes: inputting the first feature data, the user feature data, and the video feature data of each first video into the pre-trained video recommendation model. The first feature data is feature data obtained through functional interaction, for example, the first feature data is feature data obtained through functional interaction when the target user is currently using the short video platform.

[0081] In implementation, the computer device may also obtain the first feature data when a preset condition is met. The preset condition may be receiving a trigger instruction corresponding to a first-level page from a target user, or reaching a preset period. The first feature data includes one or more of holidays, time, place, season, and weather. The first feature data may be obtained by interacting with multiple functions, for example, obtaining holidays through a calendar function, obtaining locations through a positioning function, and so on. In this way, after the computer device obtains the first feature data, user feature data, and video feature data of each first video, it may input the first feature data, user feature data, and video feature data of each first video into a pre-trained video recommendation model. In this way, videos may be recommended to users in combination with the scene in which the user is currently located, so that the recommended videos match the user's current scene, thereby increasing the user's interest in the recommended videos and improving the video recommendation effect.

[0082] In some embodiments, in step 103, according to the user behavior score and the secondary page view score corresponding to each first video, determining the recommended video corresponding to the primary page in each first video includes:

[0083] The video recommendation score of each first video is determined according to the user behavior score and the secondary page view score corresponding to each first video; among each first video, a preset number of first videos are selected in descending order of the video recommendation score and determined as the recommended videos corresponding to the primary page.

[0084] In implementation, for each first video, the computer device may calculate the video recommendation score of the first video according to the user behavior score and the secondary page view score corresponding to the first video. Optionally, the computer device may calculate the average of the user behavior score and the secondary page view score as the video recommendation score; or the computer device may perform a weighted sum of the user behavior score and the secondary page view score based on a preset weight to obtain the video recommendation score.

[0085] The computer device may sort the first videos in descending order of video recommendation scores to obtain a video sequence, and then determine the first preset number of first videos as recommended videos corresponding to the first-level page. Alternatively, the computer device may also determine the first videos whose video recommendation scores are greater than a preset threshold as recommended videos corresponding to the first-level page. In this way, first videos with higher video recommendation scores may be selected as recommended videos, thereby improving the accuracy of video recommendations.

[0086] The present application also provides a video recommendation model training process, such as Figure 3 As shown, the following steps are included:

[0087] Step 301, obtaining training sample data and label data.

[0088] The training sample data includes user feature data of the sample user and video feature data of the third video browsed by the sample user on the primary page; the label data includes user behavior data corresponding to the third video and the number of views of the fourth video. The third video is a video on the primary page browsed by the sample user; the fourth video is a video that the sample user continues to browse on the secondary page after jumping to the secondary page based on the third video. Optionally, the user behavior data may include one or more of clicks, viewing time, completion rate, comments, and likes.

[0089] In implementation, the computer device may obtain user feature data of the sample user, and the video browsed by the sample user in the primary page (i.e., the third video). The computer device may also obtain video feature data of the third video, and behavior data performed by the sample user on the third video (i.e., user behavior data). In the case where the sample user jumps to the secondary page based on the third video, the computer device may also obtain the number of videos that the sample user continues to browse in the secondary page (i.e., the number of views of the fourth video). The sample user may be any user using the short video recommendation system, or a volunteer using the short video recommendation system. The number of sample users is at least one.

[0090] The computer device can use the user feature data of the sample user and the video feature data of the third video browsed by the sample user in the first-level page to form training sample data; and use the user behavior data corresponding to the third video and the number of views of the fourth video to form label data corresponding to the training sample data.

[0091] Step 302: Train the initial video recommendation model according to the training sample data and the label data to obtain a trained video recommendation model.

[0092] In implementation, the computer device can train the initial video recommendation model based on the training sample data and the label data to obtain a trained video recommendation model. Figure 4 .

[0093] When training the video recommendation model, this scheme adds the number of videos that sample users continue to browse on the secondary page after jumping to the secondary page based on the third video, so that the video recommendation model gives positive incentives to primary page videos that are consumed more on the secondary page, thereby reducing the recommendation of videos with low relevance to the secondary page.

[0094] In some embodiments, in step 301, the training sample data further includes second feature data; wherein the second feature data is feature data obtained through functional interaction when the sample user browses the third video in the first-level page. The second feature data includes one or more of holidays, time, location, season and weather.

[0095] Optionally, in step 301, training sample data and label data are obtained, including: in response to the sample user's triggering operation on the third video in the first-level page, user feature data, second feature data, and video feature data and user behavior data of the sample user are obtained; based on the user ID of the sample user and the video ID of the third video, in the video browsing log of the sample user, candidate videos belonging to the second-level page and associated with the user ID and the video ID are searched; based on the queried candidate videos, the number of views of the fourth video is determined.

[0096] In implementation, when a sample user performs a trigger operation on a certain video (i.e., the third video) on a primary page, the client can send a corresponding trigger instruction to the computer device, and then the computer device can obtain the user feature data of the sample user and record the user behavior data of the sample user. The user feature data can be uploaded by the user or authorized in advance. At the same time, the computer device can also obtain the second feature data corresponding to the sample user and the video feature data of the third video through functional interaction.

[0097] When a user browses a video, each video browsed by the user and related browsing information will be stored in a video browsing log. The computer device can determine the video browsing log corresponding to the sample user according to the user ID of the sample user, and then query the video browsing log for a video that belongs to a secondary page and is associated with the user ID and the video ID as a candidate video. Then, the computer device can determine the number of views of the fourth video based on the number of views of the queried candidate video.

[0098] In some embodiments, based on the queried candidate videos, the views of the fourth video are determined, including: determining the number of queried candidate videos as the views of the fourth video; or, among the queried candidate videos, determining the number of candidate videos whose viewing time is greater than a preset viewing time threshold as the views of the fourth video.

[0099] In implementation, the computer device may use the candidate videos found in the query as the fourth video, and the number of the candidate videos is the number of views of the fourth video. Alternatively, the computer device may determine the candidate videos whose viewing time is greater than a preset viewing time threshold from the candidate videos found in the query as the fourth video; and the total number of these candidate videos is the number of views of the fourth video.

[0100] It should be understood that although Figure 1 and Figure 3 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 and Figure 3 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0101] It can be understood that the same / similar parts between the various embodiments of the above method in this specification can refer to each other, and each embodiment focuses on the differences from other embodiments. For related points, please refer to the description of other method embodiments.

[0102] The present application also provides a video recommendation device, such as Figure 5 As shown, the device comprises:

[0103] The first acquisition module 510 is used to acquire user feature data of a target user and video feature data of each first video; wherein the first video is a video to be recommended corresponding to a first-level page;

[0104] The output module 520 is used to input the user feature data and the video feature data of each first video into a pre-trained video recommendation model, and output the user behavior score and the secondary page view score corresponding to each first video; wherein the secondary page view score of each first video represents the predicted number of target users who continue to browse the second video in the secondary page after jumping to the secondary page based on the first video; the second video is the recommended video corresponding to the secondary page;

[0105] The determination module 530 is used to determine the recommended video corresponding to the primary page in each first video according to the user behavior score and the secondary page view score corresponding to each first video.

[0106] As an optional implementation, the device further includes:

[0107] A second acquisition module is used to acquire first characteristic data of a target user; wherein the first characteristic data is characteristic data obtained through functional interaction;

[0108] The output module 520 is specifically used for:

[0109] The first feature data, the user feature data, and the video feature data of each first video are input into a pre-trained video recommendation model.

[0110] As an optional implementation, the determination module 530 is specifically configured to:

[0111] Determine the video recommendation score of each first video according to the user behavior score and the secondary page view score corresponding to each first video;

[0112] Among the first videos, a preset number of first videos are selected in descending order of video recommendation scores and determined as recommended videos corresponding to the first-level page.

[0113] As an optional implementation, the device further includes:

[0114] The third acquisition module is used to acquire training sample data and label data; wherein the training sample data includes user feature data of the sample user and video feature data of the third video browsed by the sample user on the primary page, and the label data includes user behavior data corresponding to the third video and the number of views of the fourth video, wherein the third video is a video on the primary page browsed by the sample user; and the fourth video is a video that the sample user continues to browse on the secondary page after jumping to the secondary page based on the third video;

[0115] The training module is used to train the initial video recommendation model according to the training sample data and the label data to obtain the trained video recommendation model.

[0116] As an optional implementation, the training sample data also includes second feature data; wherein the second feature data is feature data obtained through functional interaction when the sample user browses the third video in the first-level page.

[0117] As an optional implementation, the acquisition module is specifically used to:

[0118] In response to a trigger operation of the sample user on the third video in the first-level page, obtaining user feature data and second feature data of the sample user, and video feature data and user behavior data of the third video;

[0119] According to the user ID of the sample user and the video ID of the third video, querying the video browsing log of the sample user for candidate videos that belong to the secondary page and are associated with the user ID and the video ID;

[0120] Based on the queried candidate videos, the number of views of the fourth video is determined.

[0121] As an optional implementation, the acquisition module is specifically used to:

[0122] Determine the number of candidate videos found as the number of views of the fourth video; or,

[0123] Among the queried candidate videos, the number of candidate videos whose browsing time is greater than a preset browsing time threshold is determined as the browsing volume of the fourth video.

[0124] For the specific definition of the video recommendation device, please refer to the definition of the video recommendation method above, which will not be repeated here. Each module in the above video recommendation device can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor in the electronic device in the form of hardware, or can be stored in the memory of the electronic device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0125] In one embodiment, an electronic device is provided, such as Figure 6 As shown, it includes a processor 601, a communication interface 602, a memory 603 and a communication bus 604, wherein the processor 601, the communication interface 602, and the memory 603 communicate with each other through the communication bus 604; the memory 603 is used to store computer programs; the processor 601 is used to implement the above-mentioned video recommendation method steps when executing the program stored in the memory 603.

[0126] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned video recommendation method are implemented.

[0127] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0128] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0129] It should also be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data for analysis, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0130] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0131] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0132] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A video recommendation method, characterized in that: The method comprises: Obtaining user feature data of a target user and video feature data of each first video; wherein the first video is a video to be recommended corresponding to a first-level page; The user feature data and the video feature data of each of the first videos are input into a pre-trained video recommendation model, and a user behavior score and a secondary page view score corresponding to each of the first videos are output; wherein the secondary page view score of each of the first videos represents the predicted number of the target user who continues to browse the second video on the secondary page after jumping to the secondary page based on the first video; the second video is a recommended video corresponding to the secondary page; According to the user behavior score and the secondary page view score corresponding to each of the first videos, a recommended video corresponding to the primary page is determined in each of the first videos.

2. The method according to claim 1, characterized in that Before inputting the user feature data and the video feature data of each of the first videos into a pre-trained video recommendation model, the method further includes: Acquire first characteristic data of the target user; wherein the first characteristic data is characteristic data obtained through function interaction; The step of inputting the user feature data and the video feature data of each of the first videos into a pre-trained video recommendation model includes: The first feature data, the user feature data, and the video feature data of each of the first videos are input into a pre-trained video recommendation model.

3. The method according to claim 1 or 2, characterized in that: The step of determining, in each of the first videos, a recommended video corresponding to the first-level page according to the user behavior score and the second-level page view score corresponding to each of the first videos, comprises: Determine a video recommendation score for each of the first videos according to the user behavior score and the secondary page view score corresponding to each of the first videos; Among the first videos, a preset number of the first videos are selected in descending order of video recommendation scores and determined as recommended videos corresponding to the first-level page.

4. The method according to claim 1, characterized in that: The method further comprises: Acquire training sample data and label data; wherein the training sample data includes user feature data of a sample user and video feature data of a third video browsed by the sample user in the primary page, and the label data includes user behavior data corresponding to the third video and the number of views of a fourth video, wherein the third video is a video in the primary page browsed by the sample user; and the fourth video is a video that the sample user continues to browse in the secondary page after jumping to the secondary page based on the third video; The initial video recommendation model is trained according to the training sample data and the label data to obtain a trained video recommendation model.

5. The method according to claim 4, characterized in that The training sample data also includes second feature data; wherein the second feature data is feature data obtained through functional interaction when the sample user browses the third video in the first-level page.

6. The method according to claim 5, characterized in that The obtaining of training sample data and label data includes: In response to a trigger operation of the sample user on the third video in the first-level page, obtaining user feature data of the sample user, the second feature data, and video feature data and user behavior data of the third video; According to the user ID of the sample user and the video ID of the third video, searching the video browsing log of the sample user for a candidate video that belongs to the secondary page and is associated with the user ID and the video ID; Based on the queried candidate videos, the number of views of the fourth video is determined.

7. The method according to claim 6, characterized in that The determining, based on the queried candidate videos, the number of views of the fourth video includes: Determine the number of candidate videos found as the number of views of the fourth video; or, Among the queried candidate videos, the number of candidate videos whose browsing time is greater than a preset browsing time threshold is determined as the browsing volume of the fourth video.

8. A video recommendation device, characterized in that: The device comprises: A first acquisition module is used to acquire user feature data of a target user and video feature data of each first video; wherein the first video is a video to be recommended corresponding to a first-level page; An output module is used to input the user feature data and the video feature data of each of the first videos into a pre-trained video recommendation model, and output a user behavior score and a secondary page view score corresponding to each of the first videos; wherein the secondary page view score of each of the first videos represents the predicted number of the target user who continues to browse the second video in the secondary page after jumping to the secondary page based on the first video; the second video is a recommended video corresponding to the secondary page; A determination module is used to determine the recommended video corresponding to the first-level page in each of the first videos according to the user behavior score and the second-level page view score corresponding to each of the first videos.

9. An electronic device, characterized in that: The invention comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; the memory is used to store computer programs; and the processor is used to implement the method described in any one of claims 1 to 7 when executing the program stored in the memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images