A video search method, product, device and medium

Abstract

The application relates to the technical field of video search, in particular to a video search method, product, device and medium. The method comprises the following steps: obtaining historical search data of a target user, and receiving a video search request containing labels sent by the target user; determining a preference index of the target user for each label according to the historical search data; screening a plurality of recommended videos containing any label in the video search request from a video library; determining a label contained in both a recommended video and the video search request for any recommended video in the plurality of recommended videos, obtaining an intersection label, and determining a recommendation index of the recommended video according to the number of the intersection label and the preference index of the target user for each intersection label; and generating a video search result list according to the respective recommendation indexes of the plurality of recommended videos, and displaying the video search result list to the target user. The application can realize personalized and accurate video recommendation, and improves the efficiency and user experience of video search.

Description

Technical Field

[0001] This application relates to the field of video search technology, and in particular to a video search method, product, device and medium. Background Technology

[0002] With the rapid development of the internet and multimedia technologies, video resources are becoming increasingly abundant, and users' demand for quickly and accurately finding the videos they need is becoming more and more urgent. Currently, video search technology has evolved from simple keyword matching to content-based video search and personalized recommendations. In the legal field, video materials are also playing an increasingly important role, such as court transcripts, legal education videos, and legal case analyses.

[0003] Current video search technologies typically rely on keyword matching, retrieving videos relevant to user search criteria from textual information such as titles and descriptions. However, for legal videos, existing search methods fail to consider the varying levels of legal expertise among users, resulting in a lack of targeted video searches and hindering personalized recommendations based on user behavior. Summary of the Invention

[0004] To address the issue of low accuracy in personalized video recommendations for users in existing technologies, this application provides a video search method, product, device, and medium.

[0005] Firstly, this application provides a video search method, employing the following technical solution: A video search method, comprising: Acquire the target user's historical search data and receive video search requests containing tags sent by the target user; Based on the historical search data, determine the target user's preference index for each tag; Filter from the video library to select multiple recommended videos that contain any of the tags in the video search request; For any one of the multiple recommended videos, determine the tags that are included in both the recommended video and the video search request to obtain the intersection tags, and determine the recommendation index of the recommended video based on the number of intersection tags and the target user's preference index for each intersection tag; Based on the recommendation index corresponding to each of the multiple recommended videos, a list of video search results is generated and displayed to the target user.

[0006] By adopting the above technical solution, historical search data of target users can be obtained, enabling an understanding of their search preferences and providing a data foundation for recommending videos that better meet their needs. Analysis of historical search data allows for the calculation of the target user's preference index for each tag, quantifying the degree of preference and making the recommendation strategy determined by the preference index more accurate. Filtering multiple recommended videos containing any tag from the video search request from the database allows for quick location of video resources relevant to the target user's needs, improving search efficiency. By comparing the tags in the recommended videos and the video search request, intersecting tags are identified, and the recommendation index is determined based on the number of intersecting tags and the target user's preference index. This considers the matching degree between user needs and recommended video content, and the recommendation index objectively reflects the degree of fit between the recommended videos and user preferences, providing an important basis for generating search results lists. This application achieves personalized and accurate video recommendations, improving video search efficiency and user experience.

[0007] In a preferred embodiment, this application can be further configured such that the historical search data includes historical search requests and historical click records. The step of determining the target user's preference index for each tag based on the historical search data includes: Based on the historical search requests, determine the number of times the target user searches for the target tag, and based on the number of searches, determine the first preference index of the target user for the target tag, wherein the target tag is any tag contained in the videos in the video library; Based on the historical click records, determine the target user's historical click videos and the tags contained in the historical click videos, and based on the tags contained in the historical click videos, determine the target user's second preference index for the target tags; Based on the first preference index and the second preference index, the preference index of the target user for the target tag is determined.

[0008] By adopting the above technical solutions, historical search data includes the target user's past search behavior and interaction records. This information can truly reflect the target user's search preferences. By counting the number of times a user searches for a certain tag, we can intuitively understand the user's attention to that tag. The more searches, the greater the user's interest in that tag. Calculating the first preference index based on the number of searches can quantify the target user's initial preference for different tags. Historical click records reflect the video content actually watched by the target user. By analyzing the tags clicked by the user in the video, we can more accurately assess the target user's preference for a certain tag. Determining the second preference index based on historical click records can supplement and improve the preference assessment based on the number of searches. The target user's preference index for the target tag is determined based on the first and second preference indices, making the preference index more accurate and comprehensive.

[0009] In a preferred embodiment, this application can be further configured such that: determining the target user's second preference index for the target tag based on the tags contained in the historically clicked video includes: From the historical clicked videos, identify a number of videos containing the target tag, and determine the initial index of the target tag in each video, where the initial index represents the degree to which the target tag summarizes the video content; Based on the tags contained in the historical clicked videos, determine the number of times the target user clicked on the target tag; Based on the initial index of the target tag corresponding to each of the several videos and the number of clicks, the second preference index of the target user for the target tag is determined.

[0010] By adopting the above technical solution and analyzing the tags in the videos clicked in the user's history, user preferences can be accurately grasped, which helps to improve the accuracy of personalized recommendations. The initial index means that different tags have different importance in describing video content. By counting the number of times users click on target tags, the attention of target users to specific tags can be quantified, which helps to more objectively evaluate the degree of preference of target users. The second preference index is determined by comprehensive evaluation, avoiding the limitations of single indicator evaluation, and making the evaluation results more comprehensive and accurate.

[0011] In a preferred embodiment, this application can be further configured to: generate a list of video search results and display it to the target user based on the recommendation index corresponding to each of the plurality of recommended videos, including: An initial video search results list is generated based on the recommendation index corresponding to each of the multiple recommended videos. Based on the historical click records, the professionalism level of the video for each click and the corresponding viewing duration are determined. The professionalism level is used to characterize the legal expertise of the video content. For any video in the historical click record, determine the ratio of the video's viewing duration to the total viewing duration to obtain the video's viewing duration ratio; Based on the professional level and viewing duration ratio of each video in the historical click records, the target user's level of legal knowledge is determined. Based on the target user's level of legal knowledge, the initial video search results list is adjusted to obtain a new video search results list, which is then displayed to the target user.

[0012] By adopting the above technical solution, an initial list of video search results is generated based on the recommendation index of each of the multiple recommended videos, ensuring the basic accuracy of the recommended videos. By analyzing the target user's historical click records, combined with the video's level of professionalism and viewing time, the target user's level of understanding of legal knowledge can be quantified. The video search results list is then adjusted according to the level of understanding, making the displayed videos more in line with the target user's actual needs and preferences. This personalized adjustment helps improve user satisfaction, reduces the time target users spend filtering through a massive amount of videos, and improves the efficiency of information acquisition.

[0013] In a preferred embodiment, this application can be further configured as follows: determining the target user's level of legal knowledge based on the professionalism level and viewing duration ratio of each video in the historical click records includes: Based on the professional level corresponding to each video in the historical click record, the videos in the historical click record are classified and a first number of videos in each category is determined, wherein videos in the same category have the same professional level. Based on the viewing duration ratio of each video in the historical click record, videos whose viewing duration ratio reaches a preset viewing duration ratio threshold are determined as completed videos, and a second number of completed videos in each category are determined. For each type of video, the first and second quantities are used to determine the target user's level of legal knowledge.

[0014] By employing the aforementioned technical solutions, videos in the target user's historical click records are categorized, and the level of professionalism and viewing time ratio of each category are considered. This allows for a deeper understanding of the target user's behavioral patterns and learning habits, helping to more accurately assess the user's grasp of legal knowledge. Categorizing videos according to their level of professionalism and counting the number of videos in each category reflects the target user's emphasis on content of different difficulty levels. At the same time, determining the number of completed videos based on the viewing time ratio quantifies the target user's actual investment and attention to the video content. Combining the first and second numbers of videos in each category allows for a comprehensive assessment of the target user's legal knowledge, improving the targeting and effectiveness of video recommendations.

[0015] In a preferred embodiment, this application can be further configured as follows: adjusting the initial video search results list based on the target user's level of legal knowledge to obtain a new video search results list and displaying it to the target user includes: Based on the target user's level of legal knowledge, several recommended videos that match that level of knowledge are determined from the initial video search results list; The recommended videos that match the user's level of understanding are moved to the top of the initial video search results list and arranged in descending order of recommendation index to obtain an adjusted video search results list, which is then displayed to the target user.

[0016] By adopting the above technical solution, the video search list can be adjusted according to the target user's level of legal knowledge. This ensures that the recommended video content is more in line with the target user's actual level, avoiding recommending videos that are too simple or too complex. Recommended videos that match the target user's level of knowledge are moved to the top of the list and sorted according to the recommendation index. This allows users to find video content that is of interest and suitable to them more quickly, saving time in filtering videos and improving the target user's satisfaction with the video search results list.

[0017] In a preferred embodiment, this application can be further configured such that the method also includes: For any video in the video library, receive the ratings from different users after watching the video for the tags contained in the video, where the ratings indicate the degree to which the tags match the video content; Based on the score of each tag in the video, update the tags contained in the video and the initial index corresponding to the tags.

[0018] By adopting the above technical solution, the system receives ratings from different users after watching videos, providing feedback on whether the tags match the video content. This helps identify and correct incorrect tags, thereby improving the accuracy and reliability of the tags. Based on the ratings for each tag, the system updates the initial index corresponding to that tag, making the initial index more accurately reflect the degree to which the tag summarizes the video content. Allowing users to rate video tags increases interaction and participation between users and the platform, helping to improve user trust and satisfaction with the platform, thereby promoting user stickiness and activity.

[0019] Secondly, this application provides a computer program product, which adopts the following technical solution: A computer program product includes a computer program that, when executed by a processor, implements the video search method as described in any of the first aspects.

[0020] Thirdly, this application provides an electronic device that adopts the following technical solution: One or more processors; Memory; At least one application, wherein the at least one application is stored in memory and configured to be executed by at least one processor, the at least one application being configured to: perform the video search method as described in any of the first aspects.

[0021] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the video search method as described in any of the first aspects.

[0022] In summary, this application includes the following beneficial technical effects: This application, by acquiring historical search data of target users, can understand their search preferences, providing a data foundation for recommending videos that better meet user needs. Through analysis of historical search data, it calculates the target user's preference index for each tag, quantifying the degree of preference and making the recommendation strategy determined by the preference index more accurate. By filtering multiple recommended videos from the database that contain any tag in the video search request, it can quickly locate video resources related to the target user's needs, improving search efficiency. By comparing the tags in the recommended videos and the video search request, it determines the intersection tags, and determines the recommendation index based on the number of intersection tags and the target user's preference index. This considers the matching degree between user needs and recommended video content, and the recommendation index objectively reflects the degree of fit between recommended videos and user preferences, providing an important basis for generating search results lists. This application achieves personalized and accurate video recommendations, improving video search efficiency and user experience. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating a video search method provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0024] The following is in conjunction with the appendix Figure 1 - Appendix Figure 2 This application will be described in further detail.

[0025] This specific embodiment is merely an explanation of this application and is not intended to limit it. After reading this specification, those skilled in the art can make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they fall within the scope of the claims of this application.

[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0028] This application provides a video search method, such as... Figure 1 As shown, the method provided in this application embodiment is executed by an electronic device, which can be a server or a terminal device. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet, laptop, desktop computer, etc., but is not limited to these. The terminal device and the server can be directly or indirectly connected via wired or wireless communication. This application embodiment does not impose any limitations on this connection. The method includes steps S101-S105, wherein: S101. Obtain the target user's historical search data and receive a video search request containing tags sent by the target user.

[0029] In this embodiment, the target user can be any user, and the historical search data includes historical search requests and historical click records. Historical search requests are the target user's search history, including tags and timestamps included each time the target user sends a search request. Historical click records include the videos clicked by the target user each time, the viewing duration after clicking the video, and the corresponding timestamp. Specifically, a database can be pre-set to store various types of videos. Each video contains several tags to describe its content. The target user inputs a video search request containing tags through a front-end interface, and the corresponding video is retrieved from the database based on the video search request.

[0030] S102. Based on historical search data, determine the target user's preference index for each tag.

[0031] In this embodiment, on one hand, the number of times the target user searches for a target tag can be determined based on the target user's historical search requests. The target tag is any tag contained in the videos in the database. Then, based on the number of searches, a first preference index for the target tag can be determined. On the other hand, the historical click records of the target user can be used to determine the historically clicked videos and the tags contained in those videos. Based on the tags contained in those videos, a second preference index for the target tag can be determined. Finally, based on the obtained first and second preference indices, the target user's preference index for each tag is determined, thus obtaining the target user's preference index for each tag.

[0032] S103. Filter out multiple recommended videos from the video library that contain any of the tags in the video search request.

[0033] Specifically, a video search request can contain one or more tags. The system iterates through the videos in the database and selects multiple videos that contain any of the tags in the video search request as recommended videos.

[0034] S104. For any one of the multiple recommended videos, determine the tags that are included in both the recommended video and the video search request, obtain the intersection tags, and determine the recommendation index of the recommended video based on the number of intersection tags and the target user's preference index for each intersection tag.

[0035] In this embodiment, any one of the multiple recommended videos can be selected as the target recommended video. The tags contained in the target recommended video are defined as a first set, and the tags contained in the video search request are defined as a second set. The intersection of the first and second sets is determined to obtain the intersection tags. The union of the first and second sets can also be determined to obtain the union tags. Furthermore, the number of intersection tags and the number of union tags are determined, and the ratio of the number of intersection tags to the number of union tags is calculated. The sum of the preference indices of the target user for each intersection tag is also calculated. The product of the ratio and the sum of the preference indices is used as the recommendation index of the target recommended video, thus obtaining the recommendation index corresponding to each of the multiple recommended videos.

[0036] S105. Based on the recommendation index of each of the multiple recommended videos, generate a list of video search results and display it to the target user.

[0037] Specifically, multiple recommended videos can be sorted from highest to lowest according to their recommendation index to generate a video search results list. The sorted video search results list is then displayed to the target user through a front-end interface, allowing the target user to select which video to watch based on information such as the video title, description, and thumbnail in the list.

[0038] One possible implementation of this application embodiment includes: For any video in the video library, retrieve the text content contained in the video; Keyword extraction is performed on the text content to obtain multiple keywords contained in the text content, and these multiple keywords are used as tags contained in the video.

[0039] In this embodiment, speech recognition technology can be used to convert the audio content in the video into text. Subtitles or static text in the video can be extracted using optical character recognition technology to obtain the text content contained in the video. Furthermore, text cleaning and keyword extraction can be performed on the text content. Keyword extraction can be performed using any of the following methods: TF-IDF-based keyword extraction, TextRank-based keyword extraction, or deep learning-based keyword extraction (using deep learning models such as RNN, LSTM, Transformer, etc.). This embodiment does not impose specific limitations. Through keyword extraction, multiple keywords corresponding to the video and the word frequency (i.e., the number of times each keyword appears in the text content) can be obtained.

[0040] This application's embodiments, by acquiring the target user's historical search data, can understand the target user's search preferences, providing a data foundation for subsequently recommending videos that better meet the user's needs. By analyzing the historical search data, the preference index of the target user for each tag can be calculated, quantifying the degree of preference and making the recommendation strategy determined by the preference index more accurate. By filtering multiple recommended videos containing any tag in the video search request from the database, video resources related to the target user's needs can be quickly located, improving search efficiency. By comparing the tags in the recommended videos and the video search request, the intersection tags are determined, and the recommendation index is determined based on the number of intersection tags and the target user's preference index. This considers the matching degree between user needs and recommended video content. The recommendation index can objectively reflect the degree of fit between recommended videos and user preferences, providing an important basis for generating a search results list. This application achieves personalized and accurate video recommendations, improving the efficiency of video search and user experience.

[0041] In one possible implementation of this application, historical search data includes historical search requests and historical click records. Based on historical search data, determine the target user's preference index for each tag, including: Based on historical search requests, determine the number of times the target user searches for the target tag. Based on the number of searches, determine the target user's first preference index for the target tag. The target tag is any tag contained in the videos in the video library. Based on historical click records, determine the target user's historical clicked videos and the tags contained in the historical clicked videos. Based on the tags contained in the historical clicked videos, determine the target user's second preference index for the target tags. Based on the first preference index and the second preference index, determine the target user's preference index for the target tag.

[0042] In this embodiment, the historical search requests contain the tags included in each search request sent by the target user. This allows us to determine the number of times the target user searches for the target tag. The search count for tags not included in the historical search requests can be set to 0. Furthermore, the number of times the target user searches for the target tag can be used as the target user's first preference index for the target tag.

[0043] Furthermore, from historical click records, we can obtain the target user's historical clicked videos and the tags contained in those videos. For target tags, we can identify several videos containing the target tag from the historical click records and determine the initial index of the target tag in each video. The initial index represents the degree to which the target tag summarizes the video content; the higher the initial index, the more representative the target tag is of the video content. Based on the tags contained in the historical clicked videos, we can determine the number of times the target user clicked the target tag. For example, if the number of historical clicked videos containing the target tag is 5, then the target user clicked the target tag 5 times. Based on the initial index of the target tag corresponding to each of the videos containing the target tag in the historical click records, we can determine the average of the initial indices of the target tag for the videos, thus obtaining the average initial index. Then, we can calculate the product of the average initial index and the number of times the target user clicked the target tag, which serves as the target user's second preference index for the target tag.

[0044] Furthermore, after determining the target user's first preference index and second preference index for the target tag, the sum of the first preference index and the second preference index can be used as the target user's preference index for the target tag.

[0045] In this embodiment, historical search data includes the target user's past search behavior and interaction records. This information can truly reflect the target user's search preferences. By counting the number of times a user searches for a certain tag, we can intuitively understand the user's attention to that tag. The more times a user searches, the greater their interest in that tag. Calculating a first preference index based on the number of searches can quantify the target user's initial preference for different tags. Historical click records reflect the video content actually watched by the target user. By analyzing the tags clicked by the user in the video, we can more accurately assess the target user's preference for a certain tag. Determining a second preference index based on historical click records can supplement and improve the preference assessment based on the number of searches. Determining the target user's preference index for the target tag based on the first and second preference indices makes the preference index more accurate and comprehensive.

[0046] One possible implementation of this application embodiment involves determining a second preference index for a target user's target tag based on tags contained in historically clicked videos, including: Identify several videos containing the target tag from historical clicked videos, and determine the initial index of the target tag in each video. The initial index represents the degree to which the target tag summarizes the video content. Based on the tags contained in historical clicked videos, determine the number of times the target user clicked on the target tag; Based on the initial index and click count of the target tags corresponding to several videos, determine the target user's second preference index for the target tags.

[0047] In this embodiment, the initial index of the target tag in each video can be determined based on the word frequency of the target tag in the video content. Specifically, for any video in the video library, during the tag generation process, multiple keywords corresponding to the video and the word frequency of each keyword (i.e., the number of times it appears in the text content) can be obtained through keyword extraction. The obtained multiple keywords can be used as the tags corresponding to the video. The minimum word frequency can be determined from the word frequencies of each tag corresponding to the video, and the initial index of the tag corresponding to the minimum word frequency is set to 1. For any other tag, the ratio of the word frequency corresponding to the tag to the minimum word frequency can be used as the initial index of the tag. The larger the initial index, the more frequently the tag appears in the video content, that is, the more it can summarize the video content.

[0048] This application embodiment analyzes the tags in the videos clicked in the user's history to accurately grasp user preferences, which helps to improve the accuracy of personalized recommendations. The initial index means that different tags have different importance in describing video content. By counting the number of times users click on target tags, the attention of target users to specific tags can be quantified, which helps to more objectively evaluate the degree of preference of target users. The second preference index is determined by comprehensive evaluation, avoiding the limitations of single indicator evaluation, and making the evaluation results more comprehensive and accurate.

[0049] One possible implementation of this application embodiment involves generating a video search results list based on the recommendation index corresponding to each of multiple recommended videos and displaying it to a target user, including: Generate an initial list of video search results based on the recommendation index of each of the multiple recommended videos. Based on historical click records, the professionalism level of each click and the corresponding viewing time are determined. The professionalism level is used to characterize the level of legal expertise of the video content. For any video in the historical click record, determine the ratio of the video's viewing time to the total viewing time to obtain the video's viewing time ratio; Based on the professional level and viewing time ratio of each video in the historical click records, determine the target user's level of legal knowledge; The initial video search results list is adjusted based on the target users' level of legal knowledge, resulting in a final video search results list that is then displayed to the target users.

[0050] In this embodiment, after obtaining the recommendation index corresponding to each of the multiple recommended videos, the videos can be arranged in descending order of recommendation index to obtain an initial video search results list. For each video in the video library, a professionalism level can be pre-set based on human experience during video upload, representing the level of legal expertise of the video content. The number of professionalism levels can be set according to actual needs, with higher levels indicating more professional video content. Furthermore, based on historical click records, the viewing time of each video click by the target user can be determined. The database also stores the attribute information corresponding to the video, including the total duration of the video. For any video in the historical click records, the ratio of the target user's viewing time of that video to the total duration of that video is calculated to obtain the viewing time ratio corresponding to that video.

[0051] This application's embodiments generate an initial video search results list based on the recommendation index corresponding to each of the multiple recommended videos, ensuring the basic accuracy of the recommended videos. By analyzing the target user's historical click records, combined with the video's professional level and viewing duration, the target user's level of mastery of legal knowledge can be quantified. The video search results list is then adjusted according to this level of mastery, making the displayed videos more in line with the target user's actual needs and preferences. This personalized adjustment helps improve user satisfaction, reduces the target user's filtering time among a massive amount of videos, and improves the efficiency of information acquisition.

[0052] One possible implementation of this application embodiment determines the target user's level of legal knowledge based on the professional level and viewing duration ratio of each video in the historical click records, including: Based on the professionalism level of each video in the historical click records, the videos in the historical click records are classified and the first number of videos in each category is determined. Videos in the same category have the same professionalism level. Based on the viewing duration ratio of each video in the historical click records, videos whose viewing duration ratio reaches the preset viewing duration ratio threshold are identified as completed videos, and the second number of completed videos in each category is determined. For each type of video, determine the target user's level of legal knowledge by considering the first and second number of videos.

[0053] In this embodiment, the preset viewing time ratio threshold can be set according to actual needs. Optionally, the preset viewing time ratio threshold can be set to 80%. Any category of video after classification is taken as the target category video. The first quantity corresponding to the target category video represents the number of videos of the target category clicked by the target user in the historical click records. The first quantity corresponding to the target category video also represents the number of videos that the target user completed watching out of the first quantity. The ratio of the second quantity to the first quantity is calculated to obtain the viewing completion rate after clicking the target category video. The total number of each video in the historical click records is determined, and the ratio of the first quantity of the target category video to the total number of each video is determined to obtain the proportion of the target category video in each video. The product of the viewing completion rate and the proportion of the target category video is calculated as the target user's level of understanding of the target category video. After determining the target user's level of understanding of each type of video, the video with the highest level of understanding can be selected from each type of video, and the professional level of that type of video can be used as the target user's level of understanding of the law.

[0054] This application embodiment categorizes videos from the target user's historical click records, considering the professional level and viewing time ratio of each category. This allows for a deeper understanding of the target user's behavioral patterns and learning habits, helping to more accurately assess the user's grasp of legal knowledge. Categorizing videos according to professional level and counting the number of videos in each category reflects the target user's emphasis on content of different difficulty levels. At the same time, determining the number of completed videos based on the viewing time ratio quantifies the target user's actual investment and attention to the video content. Combining the first and second counts of videos in each category allows for a comprehensive assessment of the target user's legal knowledge, improving the targeting and effectiveness of video recommendations.

[0055] One possible implementation of this application embodiment involves adjusting the initial video search results list based on the target user's level of legal knowledge, obtaining a new video search results list, and displaying it to the target user, including: Based on the target user's level of legal knowledge, select several recommended videos from the initial video search results list that match their level of knowledge. Several recommended videos that match the user's level of understanding are moved to the top of the initial video search results list and arranged in descending order of recommendation index to obtain the adjusted video search results list, which is then displayed to the target user.

[0056] In this embodiment, based on the target user's level of legal expertise corresponding to their level of legal knowledge, it can be determined from a specific number of videos at the top of the initial video search results list whether videos matching the target user's level of legal expertise are included. The specific number can be set according to actual needs; optionally, it can be the top 10 videos in the list. If no videos are found, the initial video search results list is not adjusted; if they are found, several recommended videos matching the user's level of legal expertise are obtained.

[0057] For example, suppose four recommended videos that match the level of mastery are identified from the first ten videos in the initial video search results list. These four videos are then arranged in descending order of recommendation index and moved to positions 1-4 in the initial video search results list. The other six videos in the first ten videos can be arranged in descending order of recommendation index and placed in positions 5-10 in the initial video search results list, thus obtaining the adjusted video search results list.

[0058] This application's embodiments adjust the video search list based on the target user's level of legal knowledge. This ensures that the recommended video content is more in line with the target user's actual level, avoiding recommending overly simple or overly complex videos. Recommended videos that match the target user's level of knowledge are moved to the top of the list and sorted according to a recommendation index. This allows users to find interesting and suitable video content more quickly, saving time in video filtering and improving the target user's satisfaction with the video search results list.

[0059] One possible implementation of this application embodiment includes: For any video in the video library, receive the ratings of different users on the tags contained in the video after watching the video. The ratings indicate the degree to which the tags match the video content. Based on the score of each tag in the video, update the tags contained in the video and the initial index corresponding to each tag.

[0060] In this embodiment, an operation area for user rating can be set on the viewing page, providing multiple tags. Clicking a tag allows users to rate it, with ratings including "compliant," "neutral," and "non-compliant." For any video in the database containing any tag, all ratings for that tag from different users are obtained, and the proportions of "compliant," "neutral," and "non-compliant" results among all ratings are calculated. The result with the highest proportion is then determined. In one possible scenario, if the proportion of "compliant" results is the highest, the initial index corresponding to that tag in the video is increased. In another possible scenario, if the proportion of "neutral" results is the highest, the initial index corresponding to that tag in the video is not adjusted. In yet another possible scenario, if the proportion of "non-compliant" results is the highest, it is determined whether the proportion of "non-compliant" results exceeds a preset proportion threshold. If it does not exceed the threshold, the initial index corresponding to that tag in the video is decreased; if it does exceed the threshold, the tag corresponding to that tag in the video is deleted. The difference between increasing and decreasing the initial index and the preset proportion threshold can be set based on practical experience, and this embodiment does not impose specific limitations. Optionally, the preset proportion threshold can be 90%.

[0061] This application embodiment receives ratings from different users after watching videos, providing feedback on whether the tags match the video content. This helps identify and correct incorrect tags, thereby improving the accuracy and reliability of the tags. Based on the rating results of each tag, the initial index corresponding to the tag is updated, making the initial index more accurately reflect the degree to which the tag summarizes the video content. Allowing users to rate video tags increases the interaction and participation between users and the platform, helping to improve user trust and satisfaction with the platform, thereby promoting user stickiness and activity.

[0062] This application provides a computer program product, including a computer program that, when executed by a processor, implements the content shown in the aforementioned video search method embodiment.

[0063] This application provides an electronic device, such as... Figure 2 As shown, Figure 2 The illustrated electronic device 200 includes a processor 201 and a memory 203. The processor 201 and the memory 203 are connected, for example, via a bus 202. Optionally, the electronic device 200 may also include a transceiver 204. It should be noted that in practical applications, the transceiver 204 is not limited to one type, and the structure of this electronic device 200 does not constitute a limitation on the embodiments of this application.

[0064] Processor 201 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 201 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0065] Bus 202 may include a pathway for transmitting information between the aforementioned components. Bus 202 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 202 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 2 The symbol is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0066] The memory 203 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0067] The memory 203 stores the application code that executes the solution of this application, and its execution is controlled by the processor 201. The processor 201 executes the application code stored in the memory 203 to implement the content shown in the aforementioned video search method embodiment.

[0068] Figure 2 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0069] This application provides a computer-readable storage medium storing a computer program that, when run on a computer, enables the computer to execute the content shown in the aforementioned video search method embodiment.

[0070] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0071] The above are only some embodiments of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A video search method, characterized in that, include: Acquire the target user's historical search data and receive video search requests containing tags sent by the target user; Based on the historical search data, determine the target user's preference index for each tag; Filter from the video library to select multiple recommended videos that contain any of the tags in the video search request; For any one of the multiple recommended videos, determine the tags that are included in both the recommended video and the video search request to obtain the intersection tags, and determine the recommendation index of the recommended video based on the number of intersection tags and the target user's preference index for each intersection tag; Based on the recommendation index corresponding to each of the multiple recommended videos, a list of video search results is generated and displayed to the target user; The historical search data includes historical search requests and historical click records. The step of determining the target user's preference index for each tag based on the historical search data includes: Based on the historical search requests, determine the number of times the target user searches for the target tag, and based on the number of searches, determine the first preference index of the target user for the target tag, wherein the target tag is any tag contained in the videos in the video library; Based on the historical click records, determine the target user's historical click videos and the tags contained in the historical click videos, and based on the tags contained in the historical click videos, determine the target user's second preference index for the target tags; Based on the first preference index and the second preference index, determine the target user's preference index for the target tag; The step of determining the target user's second preference index for the target tag based on the tags contained in the historically clicked videos includes: From the historical clicked videos, identify a number of videos containing the target tag, and determine the initial index of the target tag in each video, where the initial index represents the degree to which the target tag summarizes the video content; Based on the tags contained in the historical clicked videos, determine the number of times the target user clicked on the target tag; Based on the initial index of the target tag corresponding to each of the videos and the number of clicks, a second preference index for the target user on the target tag is determined; Based on the recommendation index corresponding to each of the multiple recommended videos, a list of video search results is generated and displayed to the target user, including: An initial video search results list is generated based on the recommendation index corresponding to each of the multiple recommended videos. Based on the historical click records, the professionalism level of the video for each click and the corresponding viewing duration are determined. The professionalism level is used to characterize the legal expertise of the video content. For any video in the historical click record, determine the ratio of the video's viewing duration to the total viewing duration to obtain the video's viewing duration ratio; Based on the professional level and viewing duration ratio of each video in the historical click records, the target user's level of legal knowledge is determined. Based on the target user's level of legal knowledge, the initial video search results list is adjusted to obtain a new video search results list, which is then displayed to the target user.

2. The video search method according to claim 1, characterized in that, The determination of the target user's level of legal knowledge based on the professional level and viewing duration ratio of each video in the historical click records includes: Based on the professional level corresponding to each video in the historical click record, the videos in the historical click record are classified and a first number of videos in each category is determined, wherein videos in the same category have the same professional level. Based on the viewing duration ratio of each video in the historical click record, videos whose viewing duration ratio reaches a preset viewing duration ratio threshold are determined as completed videos, and a second number of completed videos in each category are determined. For each type of video, the first and second quantities are used to determine the target user's level of legal knowledge.

3. The video search method according to claim 1, characterized in that, The step of adjusting the initial video search results list based on the target user's level of legal knowledge, and then displaying the resulting video search results list to the target user, includes: Based on the target user's level of legal knowledge, several recommended videos that match that level of knowledge are determined from the initial video search results list; The recommended videos that match the user's level of understanding are moved to the top of the initial video search results list and arranged in descending order of recommendation index to obtain an adjusted video search results list, which is then displayed to the target user.

4. The video search method according to claim 1, characterized in that, The method further includes: For any video in the video library, receive the ratings from different users after watching the video for the tags contained in the video, where the ratings indicate the degree to which the tags match the video content; Based on the score of each tag in the video, update the tags contained in the video and the initial index corresponding to the tags.

5. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the steps of the video search method according to any one of claims 1 to 4.

6. An electronic device, characterized in that, include: At least one processor; Memory; At least one application, wherein the at least one application is stored in memory and configured to be executed by at least one processor, said at least one application being configured to: perform the video search method according to any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed in the computer, the computer is instructed to perform the video search method according to any one of claims 1-4.

Images