Multimedia content recommendation method, electronic device, medium and computer program product

Abstract

The embodiment of the invention discloses a multimedia content recommendation method, electronic equipment, a medium and a computer program product, and the method comprises the steps: determining a target athlete based on at least two pieces of first multimedia content watched by a user; determining a first number of each preset tag in the media tags of the at least two first multimedia contents; determining a second number of each preset tag in the media tags of the at least two second multimedia contents; each second multimedia content comprises a target athlete; the media label of each first multimedia content and the media label of each second multimedia content are part of a preset label; each preset label is used for labeling the feature of one motion quality in the multimedia content; determining the priority of each preset tag according to the number difference value between the first number and the second number of each preset tag; wherein the number difference value is in negative correlation with the priority; and based on the priority of each preset tag, determining a recommendation sequence of the to-be-recommended multimedia contents.

Description

Technical Field

[0001] This application belongs to the field of artificial intelligence technology, and specifically relates to a multimedia content recommendation method, electronic device, medium, and computer program product. Background Technology

[0002] With the accumulation of user data and the optimization of algorithm models, current recommendation systems can recommend multimedia content based on users' historical behavior. This type of recommendation method is widely used in multimedia content recommendations, such as sports events, by analyzing users' viewing records and interaction behaviors to build personalized recommendation models and improve user engagement. However, this method is prone to the "information cocoon" effect, where users only receive multimedia content that highly matches their existing interests, ignoring other potentially interesting and diverse multimedia content. This limitation restricts the breadth and depth of the recommendation system, hindering users' comprehensive understanding of the event content. Summary of the Invention

[0003] This application provides a multimedia content recommendation method, an electronic device, a medium, and a computer program product.

[0004] This application provides a multimedia content recommendation method, the method including: Based on at least two primary multimedia content pieces viewed by the user, the target athlete that the user is most interested in is identified; In the media tags of the at least two first multimedia content items, a first quantity of each preset tag in at least two preset tags is determined; in the media tags of the at least two second multimedia content items, a second quantity of each preset tag is determined; wherein, each of the at least two second multimedia content items includes the target athlete; the media tags of the at least two first multimedia content items and the media tags of the at least two second multimedia content items are part of the at least two preset tags; each preset tag is used to label a characteristic of an athletic quality included in the multimedia content; The priority of each preset tag is determined by the difference between the first quantity and the second quantity of each preset tag; wherein the difference in quantity is negatively correlated with the priority. The recommendation order of each multimedia content to be recommended is determined based on the priority of the media tags of each multimedia content to be recommended among at least two multimedia content to be recommended.

[0005] In some embodiments, before determining the first number of each preset tag among the at least two preset tags in the media tags of the at least two first multimedia contents, the method further includes: determining the characteristics of the athletic qualities included in each of the at least two first multimedia contents; determining the preset tags corresponding to the characteristics of the athletic qualities included in each of the at least two preset tags; and using the preset tags corresponding to the characteristics of the athletic qualities included in each of the first multimedia contents as the media tags of each of the first multimedia contents to obtain the media tags of the at least two first multimedia contents.

[0006] As can be seen, the method in this embodiment can obtain media tags of multimedia content based on preset tags, and match the media tags with preset tags, which helps to improve the accuracy of priority judgment and further enhance the relevance and effectiveness of multimedia content recommendation.

[0007] In some embodiments, before determining the second number of each preset tag among the media tags of at least two second multimedia contents, the method further includes: determining one or more sub-contents containing the target athlete among the at least two second multimedia contents; determining the athletic quality characteristics of the target athlete in each sub-content of the one or more sub-contents; determining preset tags corresponding to the athletic quality characteristics of the target athlete among the at least two preset tags; and using the preset tags corresponding to the athletic quality characteristics of the target athlete as media tags of the at least two second multimedia contents.

[0008] As can be seen, through the method of this embodiment, the second quantity can more accurately reflect the target athlete's mastery of various athletic qualities, which further helps to improve the accuracy of the quantity difference and the accuracy of the recommendation order.

[0009] In some embodiments, the method further includes: determining a first total value based on the sum of a first number of each preset tag in the at least two preset tags; determining a second total value based on the sum of a second number of each preset tag in the at least two preset tags; determining a comprehensive difference based on the difference between the first total value and the second total value; and displaying the comprehensive difference and the difference in the number of each preset tag on a display interface.

[0010] As can be seen, the first total value reflects the user's overall attention to various athletic qualities; the higher the user's attention, the higher their skill level in that specific athletic quality. The second total value reflects the target athlete's skill level in each athletic quality. By calculating the difference between the first and second total values, a comprehensive difference is obtained, which is then displayed in conjunction with the difference in the number of each preset tag. This provides users with the difference between their skill level and that of their most followed target athlete, as well as the difference between their skill level in each athletic quality and that of the target athlete in each athletic quality. This helps users understand their own skill level in different athletic qualities, understand the multimedia content recommendation logic, enhance the interactive experience, and facilitate users' exploration of diverse multimedia content.

[0011] In some embodiments, the method further includes: after displaying the multimedia content to be recommended, reducing the priority of the media tags of the displayed multimedia content to be recommended.

[0012] It can be seen that when a user watches a multimedia content to be recommended, by automatically lowering the priority of the media tag corresponding to that multimedia content, the probability of repeatedly recommending the same type of multimedia content can be reduced, promoting the user's exposure to more diverse multimedia content, breaking the "information cocoon", and realizing the dynamic adjustment of the recommendation strategy.

[0013] In some embodiments, before determining the target athlete that the user is most interested in based on at least two first multimedia content pieces viewed by the user, the method further includes: determining a first similarity corresponding to each multimedia content piece viewed by the user; the first similarity represents the similarity between the user's behavioral characteristics during the viewing of each multimedia content piece and preset characteristics; determining the at least two first multimedia content pieces among the multimedia content pieces viewed by the user; wherein, the first similarity corresponding to each of the at least two first multimedia content pieces is greater than a similarity threshold.

[0014] It can be seen that before identifying the target athlete, first judging which multimedia content has high reference value by the first similarity between user behavior characteristics and preset characteristics, and filtering out at least two first multimedia content, is conducive to improving the accuracy of target athlete identification and the accuracy of the first number, and improving the reliability of multimedia content recommendation.

[0015] In some embodiments, determining the target athlete that the user is most interested in based on at least two first multimedia content pieces viewed by the user includes: determining one or more target behavioral features whose second similarity is greater than a similarity threshold based on a second similarity between the user's behavioral features during the viewing of the at least two first multimedia content pieces and preset features; marking the athletes in the first multimedia content pieces corresponding to each of the one or more target behavioral features; and determining the target athlete based on the number of marks for each athlete included in the at least two first multimedia content pieces.

[0016] It can be seen that by filtering out target behavioral features through the second similarity between behavioral features and preset features, and then marking and statistically analyzing athletes based on these target behavioral features, the target athletes that users are most interested in can be determined. This enhances the ability to capture multimedia content that users are interested in and helps to improve the accuracy of identifying target athletes.

[0017] This application provides an electronic device, which includes a processor and a memory for storing computer programs capable of running on the processor; wherein, The processor is used to run the computer program to perform any of the above-described multimedia content recommendation methods.

[0018] This application provides a computer storage medium storing a computer program that, when executed by a processor, implements any of the above-described multimedia content recommendation methods.

[0019] This application provides a computer program product, including a computer program that, when executed by a processor, implements any of the above-described multimedia content recommendation methods.

[0020] This application provides a multimedia content recommendation method, electronic device, medium, and computer program product. First, based on the first multimedia content viewed by the user, the most interested target athlete is determined. The occurrence counts of preset tags in the first multimedia content and the second multimedia content containing the target athlete are then counted, yielding a first quantity and a second quantity. Next, the priority of each preset tag is determined by calculating the difference between the first and second quantities. This reflects the user's skill level in different athletic qualities and the differences in skill level among the target athletes. By determining that priority is negatively correlated with the quantity difference, multimedia content corresponding to athletic qualities where the target athlete has a high skill level but the user has a low skill level can be preferentially recommended to the user. This allows the user to view multimedia content related to athletic qualities where their skill level is low, more accurately identifying athletic qualities that the user has not fully experienced but is potentially interested in for multimedia content recommendation. This avoids the "information cocoon" effect, improves the personalization and diversity of multimedia content recommendations, and helps users comprehensively understand various competition content. Attached Figure Description

[0021] Figure 1 A flowchart of a multimedia content recommendation method provided in this application embodiment; Figure 2 A flowchart of a competition content recommendation method provided in this application embodiment; Figure 3 A schematic diagram of the structure of a multimedia content recommendation device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the composition structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0022] To ensure that recommended multimedia content (such as sports events) meets users' viewing needs, current methods primarily focus on users' desired multimedia content, using recommendation algorithms based on user data to suggest content they like. This approach can easily immerse users in an "information cocoon" created by the algorithm, causing them to miss potentially relevant but unrecommended content. Furthermore, multimedia content platforms' reliance on existing recommendation algorithms and user data can lead to stagnation in related technologies, making it difficult to meet the increasingly diverse needs of users.

[0023] The "information cocoon" effect caused by the aforementioned recommendation algorithm leads users to watch a large amount of similar and homogeneous sports content recommended by the algorithm that they prefer. This increases users' knowledge of a particular sport and the amount of similar content they acquire, giving them a sense of self-satisfaction from "watching it all, watching it to their heart's content," and making them mistakenly believe that they have understood and mastered all the content of this sport, while ignoring other aspects of the competition.

[0024] Current recommendation algorithms can only provide viewing services for event content, and cannot break free from the constraints of the recommendation mechanism. They are unable to undertake the recommendation and viewing of more comprehensive event content. In other words, there is no connection between "event content recommendation, user viewing of events, viewing needs and content characteristics" in the same dimension.

[0025] To overcome the problems existing in related technologies and achieve diversity in multimedia content recommendations, embodiments of this application provide a multimedia content recommendation method, electronic device, medium, and computer program product. The multimedia content recommendation method provided in this application can recommend more diverse multimedia content to users, thereby assisting users in increasing their attention to different types of multimedia content.

[0026] The embodiments of this application will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the embodiments provided herein are merely illustrative of the embodiments of this application and are not intended to limit the embodiments of this application. Furthermore, the embodiments provided below are some embodiments for implementing this application, and not all embodiments for implementing this application. Unless otherwise specified, the technical solutions described in the embodiments of this application can be implemented in any combination.

[0027] It should be noted that, in the embodiments of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a method or apparatus that includes a list of elements includes not only the elements expressly described, but also other elements not expressly listed, or elements inherent to implementing the method or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of other related elements in the method or apparatus that includes that element (e.g., steps in the method or units / modules in the apparatus; for example, units / modules in the apparatus may be portions of circuitry, processors, programs, or software, etc.).

[0028] The multimedia content recommendation method provided in this application includes a series of steps, but the multimedia content recommendation method provided in this application is not limited to the steps described. Similarly, the multimedia content recommendation device provided in this application includes a series of modules, but the device provided in this application is not limited to the modules explicitly described, and may also include modules that need to be set up for obtaining relevant information or processing based on information.

[0029] This application provides a multimedia content recommendation method, such as... Figure 1 As shown, Figure 1 A flowchart of a multimedia content recommendation method is shown. Figure 1 The multimedia content recommendation methods shown include: Step 101: Based on at least two primary multimedia content views, identify the target athlete that the user is most interested in.

[0030] In this embodiment of the application, the at least two first multimedia contents viewed by the user can be various videos, live broadcasts, text and image content that the user has viewed. Specifically, the at least two first multimedia contents can be sports-related content containing athletes, such as live broadcasts of events, event videos, event text and image information, etc.

[0031] The target athlete can be the athlete who appears most frequently and receives the most attention among at least two primary multimedia content pieces viewed by the user. User attention can be reflected through user behavioral characteristics, such as repeatedly watching a video or a section of text and images, taking screenshots, liking, and commenting. By analyzing user behavioral characteristics, the target athlete with the highest user attention can be identified among at least two primary multimedia content pieces.

[0032] For example, to accurately identify target athletes, technologies such as Natural Language Processing (NLP), Optical Character Recognition (OCR), and video analysis can be used to extract athletes from at least two sets of first multimedia content. Athletes are then ranked according to their frequency of appearance in each set of first multimedia content, and the athletes with the highest rankings are selected as the final identified target athletes. Furthermore, some athletes may appear frequently in each set of first multimedia content that users have saved, shared, or paid to watch; these frequently appearing athletes can also be included in the target athlete identification results.

[0033] This method records at least two pieces of primary multimedia content viewed by the user and extracts the target athlete from these two pieces of content. This allows for the accurate identification of the target athlete that the user is most interested in. It helps to analyze the sports that the user may be interested in, or the characteristics of the athletic qualities involved in the sports that the user may be interested in, thereby providing a basis for subsequent personalized recommendations.

[0034] Step 102: In the media tags of at least two first multimedia content, determine a first number of each preset tag in at least two preset tags; in the media tags of at least two second multimedia content, determine a second number of each preset tag.

[0035] In this context, each of the at least two second multimedia contents includes a target athlete; the media tags of the at least two first multimedia contents and the media tags of the at least two second multimedia contents are part of at least two preset tags; each preset tag is used to label a characteristic of a sport included in the multimedia content.

[0036] Since the characteristics of a type of multimedia content are usually fixed, for example, when multimedia content includes sports-related multimedia content, the characteristics of the sports qualities involved in different types of sports can be predetermined. Here, sports qualities include, but are not limited to, one or more of speed, strength, endurance, agility, skill, and balance. Speed ​​ability, strength ability, endurance ability, agility ability, skill ability, and balance ability can be collectively referred to as sports quality abilities.

[0037] For example, in long-distance running, the athletic qualities it reflects can be endurance and strength; in weightlifting, the athletic qualities it reflects can be strength, coordination, and explosive power; and in table tennis, the athletic qualities it reflects can be speed, balance, and agility.

[0038] Media tags can be tags that reflect athletic qualities extracted from each multimedia content, and each of the at least two first multimedia contents and each of the at least two second multimedia contents can be a part of each multimedia content.

[0039] Media tags for at least two primary multimedia content items and at least two secondary multimedia content items can be automatically generated through methods such as audio parsing, image recognition, and text recognition. For example, it can identify the sports activities included in each multimedia content item and directly determine the media tags for at least two primary multimedia content items and at least two secondary multimedia content items based on these sports activities. When a multimedia content item is identified to include long-distance running, the characteristics of the sports qualities included in the multimedia content can be directly determined based on the athletic abilities demonstrated by long-distance running, resulting in the corresponding media tags, including endurance and strength.

[0040] The first quantity for each preset tag represents the total number of times each preset tag appears in at least two first multimedia content pieces. For example, if the media tag "speed" appears 3 times in at least two first multimedia content pieces, then the first quantity corresponding to the preset tag "speed" is 3; if the media tag "endurance" appears 1 time, then the first quantity corresponding to the preset tag "endurance" is 1.

[0041] The second quantity for each preset tag represents the total number of times each preset tag appears in at least two second multimedia content pieces related to the target athlete. Each second multimedia content piece represents multimedia content containing the target athlete; for example, each second multimedia content piece could be a short video of the target athlete's competition selected from a multimedia content library.

[0042] It can be seen that at least two second multimedia content pieces may contain the same multimedia content as at least two first multimedia content pieces. In order to ensure that the media tags of at least two second multimedia content pieces can accurately reflect the target athlete's ability in each sport quality, that is, to reflect the target athlete's skill mastery of each sport quality, the total number of second multimedia content pieces can meet the preset quantity requirement, and the total number of second multimedia content pieces can be greater than the total number of first multimedia content pieces.

[0043] Furthermore, after identifying the target athlete, sub-content can be extracted from each piece of second multimedia content based on the segments in which the target athlete appears. For example, sub-content can be extracted from the second multimedia content based on consecutive image frames of the target athlete, or sub-content can be extracted from the second multimedia content based on a preset duration or a preset string length.

[0044] In this step, the first quantity includes the cumulative number of occurrences of each preset tag (such as speed) extracted from the first multimedia content. The first quantity reflects the user's attention to each athletic quality during the viewing of the first multimedia content. Attention reflects the user's understanding of each specific athletic quality. Therefore, the first quantity of each preset tag can serve as the user's digital avatar ability value in the following embodiments, reflecting the user's skill mastery of each athletic quality. The second quantity includes the cumulative number of occurrences of each preset tag (such as speed) extracted from the second multimedia content where the target athlete appears. The second quantity reflects the target athlete's performance in each athletic quality. Therefore, the second quantity of each preset tag can serve as the target athlete's digital avatar ability value in the following embodiments, reflecting the athlete's skill mastery of different athletic qualities.

[0045] Table 1 below presents the analysis methods corresponding to different athletic qualities. In practical applications, this can be achieved by processing at least two first multimedia content pieces and at least two second multimedia content pieces through audio parsing, image recognition, and text recognition. Figure 1 The given analysis method identifies features of motion qualities included in at least two first multimedia content pieces and at least two second multimedia content pieces. Here, motion qualities can be considered as immersive technology items in the embodiments below.

[0046] Table 1

[0047] By combining the analysis methods given in Table 1, each first multimedia content is tagged, resulting in at least two media tags for the first multimedia content, thus obtaining the user's tags for the immersive technology project. For example, if a video is a summary of the mistakes of Chinese players in the men's singles group stage of a sports meet, and analysis shows that this multimedia content demonstrates athletic qualities such as balance, speed, and agility, then this multimedia content needs to be tagged with three tags. Similarly, each second multimedia content is tagged, resulting in at least two media tags for the second multimedia content.

[0048] Step 103: Determine the priority of each preset tag by the difference between the first and second quantities of each preset tag; wherein the difference in quantity is negatively correlated with the priority.

[0049] Priority is used to determine the display order of recommended multimedia content. Specifically, when the difference in the number of preset tags is less than 0, it indicates that the user's skill level in the corresponding athletic quality is lower than that of the target athlete. The smaller the difference, the lower the user's attention to the target athlete's skill in that athletic quality, and the worse their skill level. Therefore, when the difference is negative, preset tags with smaller differences are given higher priority to increase the perceived skill level of the athletic quality that the user is less proficient in. It can be seen that since the difference in number is also based on the user's most concerned target athlete's skill level, the recommended multimedia content is more likely to be noticed by the user.

[0050] When the difference in the number of preset tags is equal to 0, it means that the user's skill level of the athletic quality corresponding to the preset tag is comparable to that of the athlete. Therefore, the priority of setting this preset tag can be lower than that of preset tags with a difference in the number of tags less than 0.

[0051] When the difference in the number of preset tags is greater than 0, it indicates that the user's mastery of the athletic skill represented by that preset tag is greater than the target athlete's mastery of that skill. In other words, the user is considered to have fully mastered the skill. In this case, the larger the difference, the higher the user's mastery of the skill, and the priority of preset tags with large differences can be reduced. In practical applications, when the difference in the number of preset tags is greater than 0, it indicates that the user has fully mastered the skill corresponding to that preset tag. In this case, it can be determined that this type of multimedia video will no longer be recommended.

[0052] As can be seen, since the target athletes are the athletes that users are most interested in, they can provide representative feedback on users' preferences for sports in multimedia content. By combining the quantity difference to determine the priority of preset tags, multimedia content corresponding to the sports skills that users currently have lower mastery levels can be recommended to them.

[0053] Step 104: Determine the recommendation order of each multimedia content to be recommended based on the priority of the media tags of each multimedia content to be recommended among at least two multimedia content to be recommended.

[0054] In this embodiment, each multimedia content further includes at least two multimedia content pieces to be recommended, and the media tags of the at least two multimedia content pieces to be recommended are part of at least two preset tags. After determining the priority of each preset tag, the priority of the media tags of each multimedia content piece to be recommended can be obtained, and the multimedia content to be recommended is displayed to the user based on the priority.

[0055] In practical applications, behavioral characteristics of users watching recommended multimedia content can be acquired to determine the accuracy of priority settings. For example, when a user follows or comments on recommended multimedia content, it is assumed that the user has a high level of interest in that content. When a user's viewing time for recommended multimedia content is less than the preset time, or when the user fast-forwards or skips the content, it is assumed that the user is not interested in that content. In this case, the media tag for the recommended multimedia content is marked, and the priority and recommendation frequency of that media tag are reduced.

[0056] This application provides a multimedia content recommendation method. By identifying the target athletes that users are most interested in, and combining the differences between the user's skill mastery of each sport and the target athlete's skill mastery of each sport, a precise and guided multimedia content recommendation is achieved. This effectively breaks the "information cocoon" problem caused by traditional recommendation algorithms and helps improve users' comprehensive understanding of the sports skills included in the multimedia content.

[0057] In practical applications, steps 101 to 104 can be implemented based on a processor, which can be at least one of the following: Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), Controller, Microcontroller, and Microprocessor.

[0058] To accurately obtain the media tag for each first multimedia content and improve the matching degree between the media tag and the preset tag, in some embodiments, before determining the first number of each preset tag among the media tags of at least two first multimedia contents, the method further includes: determining the characteristics of the athletic qualities included in each of the at least two first multimedia contents; determining the preset tag corresponding to the characteristics of the athletic qualities included in each of the at least two preset tags; and using the preset tag corresponding to the characteristics of the athletic qualities included in each of the first multimedia contents as the media tag for each first multimedia content, thereby obtaining the media tags for at least two first multimedia contents.

[0059] Athletic qualities reflect the key skill dimensions or athletic abilities that users or target athletes focus on when watching or participating in sports. In this embodiment, the characteristics of athletic qualities are extracted through analysis of user behavior characteristics and the multimedia content itself. Different multimedia content may contain one or more athletic quality characteristics; for example, a video about a table tennis match may simultaneously involve athletic quality characteristics such as speed, skill, and strength.

[0060] Preset tags are a predefined set of standardized tags used to represent different athletic qualities, as shown in Table 1 above. Based on the analysis method given in Table 1, the characteristics of the athletic qualities contained in each first multimedia video can be determined through technologies such as image recognition and video processing. Based on the characteristics of the athletic qualities, a matching preset tag can be determined from at least two preset tags, which will serve as the media tag for each first multimedia content.

[0061] To further improve the matching degree between the media tags of each second multimedia content and the preset tags, so that the media tags of each second multimedia content can accurately reflect the target athlete's skill mastery of each sport quality, in some embodiments, before determining the second number of each preset tag among the media tags of at least two second multimedia contents, the method further includes: determining one or more sub-contents containing the target athlete among at least two second multimedia contents; determining the characteristics of the target athlete's sport quality in each of the one or more sub-contents; determining the preset tags corresponding to the characteristics of the target athlete's sport quality among at least two preset tags; and using the preset tags corresponding to the characteristics of the target athlete's sport quality as the media tags of at least two second multimedia contents.

[0062] One or more sub-contents refer to video clips or text / image segments that are relevant to the target athlete and are extracted from at least two second multimedia content (such as video, text, or images).

[0063] In this embodiment, sub-content can be obtained by extracting images of multiple consecutive frames featuring the target athlete. Alternatively, keyword recognition can be performed on the second multimedia content to determine sub-content; here, keywords may include the target athlete's name, corresponding number, etc. Sub-content can also be obtained by extracting video clips containing the target athlete from the second multimedia content based on a preset duration; or sub-content can be obtained by extracting text from the second multimedia content based on a preset string length.

[0064] By extracting sub-content relevant to the target athlete from at least two secondary multimedia content sets, the analysis of the target athlete's skill mastery of corresponding athletic qualities can be more focused and efficient, reducing interference from redundant information. This improves the accuracy of data analysis, ensuring that the media tags of at least two secondary multimedia content sets accurately reflect the target athlete's skill mastery across athletic dimensions, and accurately reflect the target athlete's ability in each athletic dimension. For example, the target athlete may have the best skill mastery in speed and endurance, but a weaker mastery in balance and flexibility.

[0065] In this embodiment, by analyzing the athletic ability of the target athlete in each sub-content related to the target athlete, the athlete's ability in various athletic qualities can be comprehensively obtained, resulting in media tags for at least two second multimedia contents.

[0066] To further enhance the user experience of watching multimedia content and to help users understand the recommendation mechanism, in some embodiments, the above method further includes: determining a first total value based on the sum of the first quantities of each preset tag in at least two preset tags; determining a second total value based on the sum of the second quantities of each preset tag in at least two preset tags; determining a comprehensive difference based on the difference between the first total value and the second total value; and displaying the comprehensive difference and the quantity difference of each preset tag on the display interface.

[0067] Based on the method of the above embodiments, each preset tag corresponds to a first quantity and a second quantity. For example, when the preset tag is speed, the preset tag corresponds to a first quantity and a second quantity. When the first quantity of the speed tag is 5 and the second quantity is 10, it indicates that the target athlete has a higher level of speed skill mastery compared to the user's skill mastery when watching at least two pieces of first multimedia content, meaning that the user's skill mastery of speed-related athletic qualities needs further improvement.

[0068] By summing the first number of each preset label, a first total value is obtained, which can comprehensively reflect the user's overall mastery of various sports skills. Therefore, the first total value can be used as the comprehensive strength value of the user's digital avatar in the following embodiments.

[0069] By summing the second quantities of each preset label, a second total value is obtained, which can comprehensively reflect the target athlete's overall mastery of various athletic skills. Therefore, the second total value can be used as the overall strength value of the target athlete's digital avatar in the following embodiments.

[0070] By subtracting the second total value from the first total value to obtain the comprehensive difference, we can obtain the difference between the user's comprehensive mastery of various sports skills and the target athlete's comprehensive mastery of various sports skills.

[0071] At the same time, by using the difference in the number of each preset label, it is possible to further determine the difference between the user's skill level in each sport and the target athlete's skill level in each sport.

[0072] By displaying the overall difference and the quantity difference of each preset tag to the user, the user can understand their limitations when watching multimedia content, the bias in their attention, and their level of skill mastery of various sports qualities. Afterwards, the user can choose to restart the recommendation process based on the interactive interface, allowing the recommendation system to redetermine the order of each multimedia content to be recommended based on the method described in this application embodiment.

[0073] To further improve the accuracy of recommendations, in some embodiments, the method further includes: after displaying the multimedia content to be recommended, reducing the priority of the media tags of the displayed multimedia content to be recommended.

[0074] When a user watches recommended multimedia content, it indicates that they have viewed and learned about the athletic qualities corresponding to the media tags of the recommended multimedia videos, meaning their skill level in that athletic quality has improved. Building on this, to improve the user's skill level in other athletic qualities, the difference in the number of media tags corresponding to the multimedia content the user has watched can be updated. This updated difference should be greater than the original difference, thus lowering the priority of the media tags corresponding to the multimedia content the user has watched, allowing the user to prioritize watching recommended multimedia content corresponding to athletic qualities with lower skill levels.

[0075] In practice, the difference in the number of media tags corresponding to the multimedia content to be recommended that the user has watched can be increased by 1. By increasing the difference in the number of media tags, the user can receive more diverse multimedia content to be recommended, and the user can be guided to watch more multimedia content containing different types of sports qualities, thus achieving a more diverse and comprehensive multimedia content recommendation experience.

[0076] The method presented in this embodiment enables dynamic capture of users' skill mastery and a more flexible recommendation method.

[0077] Based on the recommendation method given in the above embodiments, in order to accurately obtain the user's skill mastery of different sports qualities in sports-related multimedia content, and further obtain an accurate first quantity, in some embodiments, before determining the target athlete that the user is most interested in based on at least two first multimedia contents viewed by the user, the method further includes: determining a first similarity corresponding to each multimedia content viewed by the user; the first similarity represents the similarity between the user's behavioral characteristics and preset characteristics during the viewing of each multimedia content; determining at least two first multimedia contents among the multimedia contents viewed by the user; wherein, the first similarity corresponding to each of the at least two first multimedia contents is greater than a similarity threshold.

[0078] In this embodiment, each multimedia content viewed by the user includes at least two first multimedia content items. While the user is viewing each multimedia content item, the user's behavioral characteristics during the viewing process can be captured through an interactive interface. These behavioral characteristics are then compared with preset characteristics to determine the first similarity corresponding to each multimedia content item viewed by the user.

[0079] Here, the preset features include one or more behavioral features that may occur when a user's level of attention to multimedia content exceeds a first threshold. For example, preset features include, but are not limited to, behaviors such as repeated viewing, jumping to rewind, taking screenshots, screen recording, liking, following, tipping, submitting comments and bullet comments, sharing, paying, and collecting. For instance, when a user jumps to rewind, it indicates that the user is interested in the multimedia content and has a high level of attention.

[0080] A higher first similarity score indicates a greater user interest in the multimedia content. Filtering out at least two truly representative multimedia content pieces based on first similarity helps reduce interference from accidental clicks or brief browsing that could influence user attention levels.

[0081] In practical applications, for a multimedia content, a user may exhibit multiple behavioral characteristics during viewing. Some of these behavioral characteristics have a first similarity greater than the similarity threshold, while others have a first similarity less than or equal to the similarity threshold. In this case, the multimedia content that includes those with a first similarity greater than the similarity threshold can be directly used as the first multimedia content.

[0082] In summary, this embodiment compares user behavior characteristics with preset characteristics to filter out the primary multimedia content that the user truly cares about. By determining the primary quantity through at least two primary multimedia content items, it helps to obtain an accurate assessment of the user's skill level in each sport.

[0083] Based on the methods described in the above embodiments, in order to further improve the accuracy of the identified target athletes, in some embodiments, the method of determining the target athlete that the user is most interested in based on at least two first multimedia content pieces viewed by the user includes: determining one or more target behavioral features with a second similarity greater than a similarity threshold based on the second similarity between the user's behavioral features during the viewing of at least two first multimedia content pieces and preset features; marking the athletes in the first multimedia content pieces corresponding to each of the one or more target behavioral features; and determining the target athlete based on the number of marks for each athlete contained in the at least two first multimedia content pieces.

[0084] After filtering out at least two primary multimedia content items with high user interest, since the primary multimedia content may also contain segments or athletes that the user is not interested in, for example, the user may skip or fast forward to segments or athletes that are not interesting when watching the primary multimedia content.

[0085] To identify the target athletes with the highest user engagement, we can determine the target behavioral characteristics within the first multimedia content viewed by the user, and then identify and tag the corresponding athletes within that content. For example, when a user comments on an athlete within the multimedia content, that athlete is tagged; or when a user rewatches a segment featuring a particular athlete, that athlete is tagged. Finally, based on the number of tags for each athlete, the athlete with the most tags is identified as the target athlete.

[0086] This application provides a multimedia content recommendation method. The following uses multimedia content as the example of competition content to further illustrate the recommendation method given in the above embodiment.

[0087] Figure 2 A flowchart of a method for recommending event content is shown, such as... Figure 2 As shown, the methods for recommending event content include: Step 201: Analyze user behavior.

[0088] By analyzing the user's interaction with the event content, such as the user's behavior related to the event content they are watching on the interactive interface, user behavior can be analyzed.

[0089] In practical applications, the event content can also be recorded based on user behaviors such as repeated viewing, rewinding, screenshotting, screen recording, liking, following, tipping, submitting comments and bullet screens, sharing, paying, and collecting. This content serves as the target event content, namely the first multimedia content in the above embodiments.

[0090] Step 202: Identify the target athlete.

[0091] While performing step 201, based on the method of the above embodiments, it is possible to analyze and define the athlete that the user is most concerned about in the event content that the user watches and reads as the target athlete, and at the same time, it is possible to determine the event content of the target athlete that includes the target athlete, that is, the second multimedia content given in the above embodiments.

[0092] Step 203: Identify immersive technology projects.

[0093] The characteristics of the technical items (i.e., one or more athletic qualities in the above embodiments) involved in each event content in the event content library can be analyzed, including audio parsing, image recognition, text recognition, etc. of each event content. The analyzed technical items can be defined as immersive technical items, resulting in immersive technical items as shown in Table 1. The corresponding tag for each immersive technical item is determined, which is the preset tag in the above embodiments.

[0094] In practical applications, during the execution of step 201, based on the method in step 201, each immersive technology item corresponding to the user and the tag of each immersive technology item corresponding to the user can be determined in the target event content, that is, the media tag of the first multimedia content in the above embodiment.

[0095] In practical applications, during the execution of step 202, all the event content of the target athlete that has appeared can be analyzed using methods such as audio parsing, image recognition, and text recognition to obtain the tag of each immersive technology item corresponding to the target athlete in all the event content of the target athlete, that is, the media tag of the second multimedia content in the above embodiment.

[0096] For example, a short video depicting a target athlete's winning moment in a table tennis match, analyzed to reveal that the athlete's performance includes attack, technique, and power, could be tagged with the following labels for each immersive technical element of the target athlete based on that short video: attack, technique, and power. In practical applications, steps 202 and 203 can be performed simultaneously.

[0097] Then, steps 204 and 205 can be executed simultaneously.

[0098] Step 204: Identify the user's digital clone.

[0099] The user digital clone includes two indicators: the overall strength value of the user digital clone and the ability value of the user digital clone.

[0100] Based on the tags for each immersive technology project corresponding to the user obtained in step 203, the number of tags for each immersive technology project is counted, that is, the first number of each preset tag in the above embodiment.

[0101] The user's digital avatar capability value is determined based on the number of tags associated with each immersive technology project. The minimum digital avatar capability value is 0. A user's digital avatar capability value of 0 indicates that the target event content does not contain the features of that immersive technology project.

[0102] The sum of the user's digital clone ability values ​​is the first total value in the above embodiment, which is used as the user's overall digital clone strength value.

[0103] Based on the method described in this step, Table 2 shows the user's digital clone ability values. It can be seen that the user's digital clone ability includes speed ability, strength ability, etc., and the range of the user's digital clone ability value is [0, +∞).

[0104] Table 2

[0105] Step 205: Identify the target athlete's digital clone.

[0106] The target athlete digital clone includes two indicators: the target athlete digital clone's overall strength value and the target athlete digital clone's ability value.

[0107] Based on the tags for each immersive technology item corresponding to the target athlete obtained in step 203, the number of tags for each immersive technology item corresponding to the target athlete is counted, which is the second number of each preset tag in the above embodiment.

[0108] Based on the number of tags for each immersive technology item corresponding to the target athlete, the digital clone ability value of the target athlete is determined. The minimum value of the digital clone ability value of the target athlete is 0, which means that the target athlete's event content does not contain the feature of that immersive technology item.

[0109] The digital clone ability values ​​of the target athlete are added together to obtain the second total value in the above embodiment, which is the overall strength value of the digital clone of the target athlete.

[0110] The range of the target athlete's digital clone ability value is [0, +∞). The target athlete's digital clone ability value can also be set with reference to Table 2.

[0111] Step 206: Determine the overall capability difference.

[0112] Based on the user's digital avatar's overall strength value and the target athlete's digital avatar's overall strength value determined in steps 204 and 205, the user's digital avatar's overall strength value is subtracted from the target athlete's digital avatar's overall strength value to obtain the overall ability difference, which is the overall difference value in the above embodiment.

[0113] When the overall ability difference is negative, the user's digital avatar is considered to be weaker than the target athlete; when the overall ability difference is positive, the user's digital avatar is considered to be stronger than the target athlete; when the overall ability difference is 0, the user's digital avatar is considered to be comparable to the target athlete.

[0114] Step 207: Determine the individual ability difference.

[0115] The user's digital clone ability value determined in steps 204 and 205 is compared with the target athlete's digital clone ability value to obtain the difference in individual abilities.

[0116] For any immersive technology project, the user's digital avatar ability value corresponding to that immersive technology project is subtracted from the target athlete's digital avatar ability value to determine the avatar ability difference for that immersive technology project. This results in the individual ability difference value for each immersive technology project. The range of values ​​for the individual ability difference value for each immersive technology project is (…). Integers in ).

[0117] In practical applications, steps 206 and 207 can be performed simultaneously.

[0118] Step 208: Analyze the differences in individual abilities to obtain a user event content recommendation scheme.

[0119] Based on the individual ability difference corresponding to each immersive technology project, the user's individual ability for each immersive technology project is analyzed, that is, the degree of technical mastery.

[0120] Determine individual ability improvement plans: When a user's digital avatar ability value in a certain immersive technology activity is lower than that of an athlete, the improvement plan is "Learning"; when a user's digital avatar ability value in a certain immersive technology activity is stronger than that of an athlete, the improvement plan is "Maintaining"; when a user's digital avatar ability value in a certain immersive technology activity is equal to that of an athlete, the improvement plan is "Enhancing". The conclusions of the individual ability improvement plans for users are represented by "Learning," "Maintaining," and "Enhancing."

[0121] The user's competition content recommendation scheme is determined based on the conclusions of the improvement scheme for each individual ability corresponding to each immersive technology project. Immersive technology projects corresponding to individual abilities with the improvement scheme of "learning" will be given primary recommendations, while immersive technology projects corresponding to individual abilities with the improvement scheme of "enhancement" will be given secondary recommendations, and immersive technology projects corresponding to individual abilities with the improvement scheme of "maintaining" can be left unrecommended.

[0122] The individual ability improvement plans are derived based on the difference in individual abilities. When a user's individual ability is weaker than the target athlete's, the improvement plan is "Learning"; when a user's individual ability is stronger than the target athlete's, the improvement plan is "Maintaining"; and when a user's individual ability is equal to the target athlete's, the improvement plan is "Strengthening". Table 3 shows the correspondence between immersive technology projects and individual ability improvement plans.

[0123] Table 3

[0124] For each individual ability improvement plan categorized as "Learning," the difference in their respective ability values ​​is ranked, with the smallest value being the first recommended plan. This process continues, with the second smallest value being the second recommended plan, until all individual ability recommendations for "Learning" are completed. Ability improvement plans categorized as "Enhancement" are recommended last.

[0125] The immersive technology projects corresponding to each individual capability that needs to be recommended will be searched for corresponding event content in the event content database and defined as event content to be recommended.

[0126] Based on the above method, Table 4 shows a recommendation method that can rank the recommended event content based on the difference in individual capabilities.

[0127] Table 4

[0128] Step 209: Present the results of the digital clone competition and recommended event content to the user.

[0129] The results obtained from steps 206, 207, and 208 are presented to the user. The overall ability difference, individual ability difference, and the recommended event content (i.e., the recommended content in Table 4) obtained from the above steps are presented to the user in the following order.

[0130] It can display whether the user's digital avatar's overall strength is weaker than the target athlete, stronger than the target athlete, or comparable to the target athlete.

[0131] The individual ability information obtained from the above steps will be presented to the user, showing which one or more of the user's digital clone ability values ​​are weaker than the target athlete, or which one or more of the user's digital clone ability values ​​are comparable to the target athlete, or which one or more of the user's digital clone ability values ​​are stronger than the target athlete.

[0132] The recommended event content obtained from the above steps is presented to the user in numbers 1 to N. After the user completes the playback and viewing, this recommended content will no longer be recommended in subsequent sessions. At the same time, the individual ability difference corresponding to this content is incremented by 1.

[0133] Users can restart step 201 at any time, recalculate and analyze according to the method of steps 201 to 208 in this embodiment, and obtain a new event content recommendation scheme based on the comprehensive ability difference and the single-item clone ability difference after restarting the calculation.

[0134] This application provides a multimedia content recommendation method. By calculating the number of times the user and the target athlete correspond to each preset tag, the order of recommending multimedia content is determined based on the number of times the user and the target athlete correspond to each preset tag. This provides a new multimedia content recommendation method, clarifies the quantification method of the differences between the user and the target athlete in various athletic qualities, and improves the accuracy of multimedia content recommendation.

[0135] This application provides a method for determining the recommendation order during the multimedia content recommendation process, and adds a priority adjustment strategy after the user views the multimedia content to be recommended, thereby improving the accuracy of subsequent recommendations.

[0136] Those skilled in the art will understand that, in the above-described method of the specific implementation, the order in which each step is written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0137] Based on the multimedia content recommendation method proposed in the foregoing embodiments, this application also provides a multimedia content recommendation device. Figure 3 A schematic diagram of a multimedia content recommendation device is shown, such as... Figure 3 As shown, the multimedia content recommendation device includes: Processing module 301 is configured to: determine the target athlete that the user is most interested in based on at least two first multimedia content pieces viewed by the user; determine a first quantity of each preset tag among at least two preset tags in the media tags of the at least two first multimedia content pieces; determine a second quantity of each preset tag among the media tags of the at least two second multimedia content pieces; wherein each of the at least two second multimedia content pieces includes the target athlete; the media tags of the at least two first multimedia content pieces and the media tags of the at least two second multimedia content pieces are part of the at least two preset tags; each preset tag is used to label a characteristic of an athletic quality included in the multimedia content; and determine the priority of each preset tag by the difference between the first quantity and the second quantity of each preset tag; wherein the difference in quantity is negatively correlated with the priority.

[0138] The recommendation module 302 is used to determine the recommendation order of each multimedia content to be recommended based on the priority of the media tags of each multimedia content to be recommended among at least two multimedia content to be recommended.

[0139] In practical applications, the processing module 301 and the recommendation module 302 can be implemented based on a processor and a communication device.

[0140] In some embodiments, before determining the first number of each preset tag among the at least two preset tags in the media tags of at least two first multimedia contents, the processing module 301 is further configured to determine the characteristics of the athletic qualities included in each of the at least two first multimedia contents; determine the preset tags corresponding to the characteristics of the athletic qualities included in each of the at least two preset tags; and use the preset tags corresponding to the characteristics of the athletic qualities included in each of the first multimedia contents as the media tags of each first multimedia content to obtain the media tags of at least two first multimedia contents.

[0141] In some embodiments, before determining the second number of each preset tag among the media tags of at least two second multimedia contents, the processing module 301 is further configured to: determine one or more sub-contents containing the target athlete among the at least two second multimedia contents; determine the characteristics of the target athlete's athletic qualities in each of the one or more sub-contents; determine the preset tags corresponding to the characteristics of the target athlete's athletic qualities among the at least two preset tags; and use the preset tags corresponding to the characteristics of the target athlete's athletic qualities as media tags of the at least two second multimedia contents.

[0142] In some embodiments, the processing module 301 is further configured to determine a first total value based on the sum of the first quantities of each preset tag in at least two preset tags; determine a second total value based on the sum of the second quantities of each preset tag in at least two preset tags; and determine a comprehensive difference based on the difference between the first total value and the second total value. The multimedia content recommendation device further includes a display module, which is configured to display the comprehensive difference and the quantity difference of each preset tag on a display interface.

[0143] In some embodiments, the recommendation module 302 is further configured to reduce the priority of the media tags of the displayed multimedia content to be recommended after displaying the multimedia content to be recommended.

[0144] In some embodiments, before determining the target athlete that the user is most interested in based on at least two first multimedia content pieces viewed by the user, the processing module 301 is further configured to determine a first similarity corresponding to each multimedia content piece viewed by the user; the first similarity represents the similarity between the user's behavioral characteristics during the viewing of each multimedia content piece and preset characteristics; at least two first multimedia content pieces are determined among the multimedia content pieces viewed by the user; wherein, the first similarity corresponding to each of the at least two first multimedia content pieces is greater than a similarity threshold.

[0145] In some embodiments, the processing module 301 is specifically configured to: determine one or more target behavioral features whose second similarity is greater than a similarity threshold based on the second similarity between the user's behavioral features during the viewing of at least two first multimedia contents and preset features; mark the athletes in the first multimedia contents corresponding to each of the one or more target behavioral features; and determine the target athlete based on the number of marks for each athlete contained in the at least two first multimedia contents.

[0146] It should be noted that the descriptions of the above device embodiments are similar to those of the above method embodiments, and have similar beneficial effects. For technical details not disclosed in the device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0147] It should be noted that, in the embodiments of this application, if the above-described methods are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a terminal, server, etc.) to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0148] This application also provides an electronic device. Figure 4 This is a schematic diagram of the composition structure of an electronic device provided in an embodiment of this application, as shown below. Figure 4 As shown, the electronic device 40 may include: Memory 401 is used to store executable instructions.

[0149] The processor 402 is configured to implement any of the multimedia content recommendation methods described above when executing the executable instructions stored in the memory 401.

[0150] The processor 402 mentioned above can be at least one of ASIC, DSP, DSPD, PLD, FPGA, CPU, controller, microcontroller, and microprocessor.

[0151] The aforementioned computer-readable storage medium or memory 401 may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM), etc.; it may also be various terminals that include one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc.

[0152] This application embodiment further provides a computer storage medium storing computer-executable instructions, which are used to implement any of the multimedia content recommendation methods provided in the above embodiments.

[0153] Correspondingly, this application embodiment further provides a computer program product, the computer program product including computer executable instructions, which are used to implement any of the multimedia content recommendation methods provided in the above embodiments.

[0154] In some embodiments, the functions or modules of the apparatus provided in this application can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0155] The description of the various embodiments above tends to emphasize the differences between the various embodiments. The similarities or similarities between them can be referred to, and for the sake of brevity, they will not be repeated here.

[0156] The methods disclosed in the various method embodiments provided in this application can be arbitrarily combined to obtain new method embodiments without conflict.

[0157] The features disclosed in the various product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0158] The features disclosed in the various method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0159] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0160] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of protection of this application, and these forms are all within the protection scope of this application.

Claims

1. A multimedia content recommendation method, characterized in that, The method includes: Based on at least two primary multimedia content pieces viewed by the user, the target athlete that the user is most interested in is identified; In the media tags of the at least two first multimedia content items, a first quantity of each preset tag in at least two preset tags is determined; in the media tags of the at least two second multimedia content items, a second quantity of each preset tag is determined; wherein, each of the at least two second multimedia content items includes the target athlete; the media tags of the at least two first multimedia content items and the media tags of the at least two second multimedia content items are part of the at least two preset tags; each preset tag is used to label a characteristic of an athletic quality included in the multimedia content; The priority of each preset tag is determined by the difference between the first quantity and the second quantity of each preset tag; wherein the difference in quantity is negatively correlated with the priority. The recommendation order of each multimedia content to be recommended is determined based on the priority of the media tags of each multimedia content to be recommended among at least two multimedia content to be recommended.

2. The method according to claim 1, characterized in that, Before determining the first number of each preset tag among the at least two preset tags in the media tags of the at least two first multimedia contents, the method further includes: Determine the characteristics of the athletic qualities included in each of the at least two first multimedia contents; Determine the preset tag corresponding to the feature of the athletic quality included in each of the at least two preset tags; By using preset tags corresponding to the characteristics of athletic qualities included in each of the first multimedia contents as media tags for each of the first multimedia contents, media tags for the at least two first multimedia contents are obtained.

3. The method according to claim 1, characterized in that, Before determining the second number of each preset tag among the media tags of at least two second multimedia contents, the method further includes: Among the at least two second multimedia contents, one or more sub-contents containing the target athlete are identified; In each of the one or more sub-contents, the characteristics of the target athlete's athletic qualities are determined; Determine the preset label corresponding to the characteristic of the target athlete's athletic quality from the at least two preset labels; The preset tags corresponding to the characteristics of the target athlete's athletic qualities are used as media tags for the at least two second multimedia content items.

4. The method according to claim 1, characterized in that, The method further includes: A first total value is determined based on the sum of the first quantities of each of the at least two preset labels; A second total value is determined based on the sum of the second quantities of each of the at least two preset labels; The comprehensive difference is determined based on the difference between the first total value and the second total value; The display interface shows the overall difference and the quantity difference of each preset label.

5. The method according to claim 1, characterized in that, The method further includes: After displaying the multimedia content to be recommended, reduce the priority of the media tags of the displayed multimedia content.

6. The method according to claim 1, characterized in that, Before determining the target athlete of greatest interest to the user based on at least two pieces of first multimedia content viewed by the user, the method further includes: Determine the first similarity for each multimedia content viewed by the user; the first similarity represents the similarity between the user's behavioral characteristics during the viewing of each multimedia content and preset characteristics; The at least two first multimedia contents are identified from the multimedia content viewed by the user; wherein the first similarity corresponding to each of the at least two first multimedia contents is greater than a similarity threshold.

7. The method according to claim 1, characterized in that, The process of determining the target athlete of greatest interest to the user based on at least two pieces of first multimedia content viewed by the user includes: Based on the second similarity between the user's behavioral characteristics during the viewing of the at least two first multimedia contents and preset characteristics, one or more target behavioral characteristics with the second similarity greater than a similarity threshold are determined; The athletes in the first multimedia content corresponding to each of the one or more target behavioral features are marked; The target athlete is determined based on the number of tags for each athlete contained in the at least two first multimedia contents.

8. An electronic device, characterized in that, The electronic device includes a processor and a memory for storing computer programs capable of running on the processor; wherein, The processor is used to run the computer program to perform the method according to any one of claims 1 to 7.

9. A computer storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images