A video recommendation method for preventing and controlling children's short video addiction

Abstract

The present invention relates to the field of artificial intelligence, and discloses a video recommendation method for preventing and controlling children's short video addiction, including: determining a first short video list that has been played within a first time window; obtaining media resource tags corresponding to each video in the short video list that has been played; determining the tag time distribution according to the media resource tags corresponding to the short video list and the playing time of the short video; determining the first media resource tag distribution according to the media resource tags and the short video list; determining the first probability that the visitor is a child according to the first media resource tag distribution and the tag time distribution; in response to the first probability exceeding a first threshold, updating the video resource list in the video push list to a second short video list; based on the user's operation on the video in the second short video list, updating the video list to a third short video list. The method of the present invention creates a healthy and educational digital entertainment environment for child users, and prevents children from indulging in short videos.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a video recommendation method for preventing and controlling children's addiction to short videos. Background Art

[0002] In the digital age, short video platforms have rapidly gained popularity due to their rich content and convenient user experience, becoming one of the main sources of daily entertainment for people. Especially for children and adolescent users, these platforms provide a virtual space full of fun and knowledge. However, as their influence continues to expand, the negative impact of short videos on child users has also begun to emerge, especially the problem of children's addiction to short videos, which has attracted widespread attention.

[0003] The current short video recommendation system mainly uses machine learning algorithms to push personalized content based on users' historical viewing data, interactive behaviors, and personal preferences. Although this algorithm can improve user experience and increase user retention time, for young children, excessive personalized recommendations may cause them to be exposed to electronic screens too much, neglect learning and physical activities, and thus affect their physical and mental health. For example, long-term screen viewing can affect children's eyesight, disrupt their biological clocks, and affect their normal life and learning rhythm.

[0004] In addition, the existing short video recommendation mechanism is still insufficient in content filtering and age-appropriateness control. Due to age restrictions, child users have weak judgment and self-control abilities and are easily influenced by inappropriate content. Existing systems often fail to effectively identify the real age and needs of child users, resulting in the recommendation system sometimes pushing inappropriate content to children, such as overly commercialized advertisements or unhealthy lifestyle guidance, which not only violates the original intention of education and guidance, but also may pose a potential threat to children's mental health.

[0005] In addition, the current system also shows limitations in terms of supervision and parental control functions. Although some platforms provide child modes or parental monitoring options, these measures often rely on parents' active settings and supervision. For parents who do not have relevant knowledge, how to effectively use these functions has become a major challenge. Therefore, how to provide a system from a technical level that can ensure the richness and fun of content while ensuring healthy use by children is an urgent problem to be solved. Summary of the invention

[0006] At least one aspect or advantage of the invention will be set forth in part in the following description, or may be obvious from the description, or may be learned by practice of the disclosed subject matter.

[0007] The present invention provides a video recommendation method for preventing and controlling children's addiction to short videos, aiming to create a safer, healthier and more educational digital entertainment environment for child users.

[0008] According to a first aspect of the present invention, a video recommendation method for preventing and controlling children's addiction to short videos comprises:

[0009] Determine a first short video list that has been played within a first time window;

[0010] Obtain the media resource tag corresponding to each video in the short video list that has been played, wherein the media resource tag includes a main tag and a sub-tag;

[0011] Determine the label time distribution according to the media resource labels corresponding to the short video list and the playing time of the short videos;

[0012] Determine a first media resource tag distribution according to the media resource tags and the short video list;

[0013] Determine a first probability that the visitor is a child according to the first media resource tag distribution and the tag time distribution;

[0014] In response to the first probability exceeding the first threshold, updating the video resource list in the video push list to a second short video list, wherein the second short video list has a second media resource tag distribution;

[0015] Based on the user's operation on the videos in the second short video list, updating the video list;

[0016] The updating of the video list includes updating the video resource list to a third short video list, wherein the third short video list has a third media resource tag distribution or a fourth media resource tag distribution;

[0017] In the fourth media resource tag distribution, the proportion of the first type of media resource tag distribution is higher than the proportion of the second type of media resource tag distribution;

[0018] In the second media resource tag distribution and the third media resource tag distribution, the proportion of the first type of media resource tag distribution is lower than the proportion of the second type of media resource tag distribution.

[0019] According to an embodiment of the present invention, the first probability is obtained based on a multi-layer perceptron, the input of the multi-layer perceptron is the first media resource label distribution and the label time distribution, and the output of the multi-layer perceptron is the probability that the user is a child.

[0020] According to an embodiment of the present invention, the tag time distribution is determined according to the main tag corresponding to each video and the playing time of the short video.

[0021] According to one embodiment of the present invention, the second short video list is obtained according to the following method:

[0022] Determine a first tag list according to the number of plays from low to high, where the number of plays is the number of plays of all videos under the tag;

[0023] Remove duplicate tags of media resource tags corresponding to each video in the first tag list and the short video list that has been played, and obtain a second tag list;

[0024] Acquire, from a database according to the second tag list, a first short video set of short videos corresponding to the second tag list;

[0025] Obtaining a second short video set from a database according to tag information of a number of short videos played last in the first short video list, wherein the number of short videos in the second short video set is less than the number of short videos in the first short video set;

[0026] A second short video list is generated based on the first short video set and the second short video set.

[0027] According to an embodiment of the present invention, the first tag list is selected from a first category of media resource tags.

[0028] According to an embodiment of the present invention, in the second short video list, the short videos included in the second short video set are rendered to an invisible area of ​​the video recommendation component.

[0029] According to one embodiment of the present invention, in response to the user's operation of returning to the main interface of the short video APP, a short video list is loaded on the main interface, and the short video list has a second media resource tag distribution.

[0030] According to one embodiment of the present invention, updating the video list includes the following steps:

[0031] In response to the selected duration of a video in the second short video list of the user exceeding the first duration threshold or the short video selected by the user being a member of the first short video set, randomly selecting a video from the second short video list and loading resources and acquiring a third short video list, wherein the third short video list has a third media resource tag distribution; or

[0032] In response to a selected duration of a video in the second short video list selected by the user being lower than the first duration threshold, and the short video selected by the user being a member of the first short video set, obtaining a third short video list, wherein the third short video list has a fourth media resource tag distribution; or

[0033] In response to a selected duration of a video in the second short video list selected by the user being lower than a first duration threshold, and the short video selected by the user being a member of the second short video set, a third short video list is obtained, wherein the third short video list has a third media resource tag distribution.

[0034] According to a second aspect of the present invention, an electronic device includes a processor and a memory; the memory is used to store a program; the processor executes the program to implement a video recommendation method for preventing and controlling children's addiction to short videos.

[0035] According to a third aspect of the present invention, a computer-readable storage medium stores a program, and the program is executed by a processor to implement a video recommendation method for preventing and controlling children's addiction to short videos.

[0036] The technical solution of the present invention includes at least the following technical effects:

[0037] By precisely controlling the types of content that children can access, the platform increases the protection of child users and prevents them from being exposed to inappropriate video content;

[0038] Using the user's viewing history and interaction data to dynamically update the video recommendation list, it can more accurately meet the user's personalized needs;

[0039] Use label time distribution and multi-layer perceptron models to accurately identify child users and adjust push lists to prevent inappropriate content from affecting children.

[0040] By adjusting the ratio of first- and second-category labels, we can increase content suitable for children and reduce the risk of addiction;

[0041] Update video lists in real time based on user operations, optimize recommended content, and improve user experience;

[0042] Use machine learning models to accurately identify child users and improve anti-addiction effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flowchart of a video recommendation method for preventing and controlling children's short video addiction provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The present disclosure will now be discussed with reference to several exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and thus implement the present disclosure, rather than implying any limitation on the scope of the present disclosure.

[0045] As used herein, the term "including" and its variants are to be interpreted as open-ended terms meaning "including but not limited to". The term "based on" is to be interpreted as "based at least in part on". The terms "one embodiment" and "an embodiment" are to be interpreted as "at least one embodiment". The term "another embodiment" is to be interpreted as "at least one other embodiment". The orientation or position relationship indicated by the terms "upper", "lower", "left", "right", "front", "back", "top", "bottom", "inner", "outer", "vertical", "horizontal", "lateral", "longitudinal" and the like is based on the orientation or position relationship shown in the accompanying drawings. These terms are mainly for better describing the present application and its embodiments, and are not used to limit the indicated devices, elements or components to have a specific orientation, or to be constructed and operated in a specific orientation. In addition, in addition to being used to indicate an orientation or position relationship, some of the above terms may also be used to indicate other meanings, such as the term "upper" may also be used to indicate a certain dependency or connection relationship in some cases. For those of ordinary skill in the art, the specific meanings of these terms in this application can be understood according to the specific circumstances. In addition, the terms "install", "set", "provided with", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral structure; it can be a mechanical connection, or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or it can be an internal connection between two devices, elements or components. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to the specific circumstances. In addition, the terms "first", "second", etc. are mainly used to distinguish different devices, elements or components (the specific types and structures may be the same or different), and are not used to indicate or imply the relative importance and quantity of the indicated devices, elements or components. Unless otherwise specified, "plurality" means two or more.

[0046] With the popularization of the Internet and smart mobile devices, short video platforms have quickly become the main way for global users, especially children and teenagers, to obtain entertainment and information. These platforms use algorithm recommendation systems to provide personalized video content based on users' historical behavior and preferences to increase user stickiness and viewing time. However, this personalized recommendation also brings about the problem of children's overuse and exposure to inappropriate content.

[0047] Existing technologies are usually unable to accurately identify the age of users, especially children, resulting in an inability to provide appropriate content filtering for children. Although some platforms try to filter content through child modes, these mechanisms often rely on parents' active settings and cannot fully cover all potentially inappropriate content. Current technical solutions also lack effective control mechanisms to limit children's viewing time and prevent children from becoming overly addicted to short videos, which affects their physical and mental health and learning development. These issues highlight the shortcomings of existing technologies in protecting child users, and there is an urgent need to provide safer and more responsible content recommendation solutions through technological innovation.

[0048] See also Figure 1 As shown, a video recommendation method for preventing and controlling children's short video addiction in this application is provided, and the method comprises:

[0049] A first short video list that has been played within a first time window is determined.

[0050] Specifically, first, the time window is set to determine the first time window, that is, the system needs to track all short videos watched by the user in the recent period of time; then the viewing data is collected, using the user's viewing history. Whenever the user starts watching a new video, the system records the video ID, the start time and end time of the viewing, and the interruption time point when the user interrupts the viewing; finally, a short video list is generated. The system extracts all videos played within the specified time window from the user's viewing history database, and generates a first short video list containing detailed information on each video, including video ID, total playback time, and number of views.

[0051] The media resource tag corresponding to each video in the short video list that has been played is obtained, and the media resource tag includes a main tag and a sub-tag.

[0052] Specifically, extract label information. For each video, extract media resource labels from the video metadata, where the main label usually indicates the main category of the video, and the secondary label provides more specific information. Establish a label index for each video in the back-end database to ensure that the label data of each video can be effectively called and processed, and associate the label data with the video ID to ensure that it can be accurately called when the recommendation algorithm is running. Develop automated scripts or use existing content management system (CMS) functions to process large amounts of video data and extract relevant labels. For videos with missing labels, you can set default labels or predict possible labels through artificial intelligence models. Regularly review and optimize the label system to ensure the accuracy and timeliness of labels, especially strictly monitor the suitability of children's content. At the same time, introduce manual review links to improve the accuracy and reliability of label classification.

[0053] The label time distribution is determined according to the media resource labels corresponding to the short video list and the playing time of the short video. In this embodiment, the label time distribution is determined according to the main labels corresponding to each video and the playing time of the short video.

[0054] Specifically, the play time of each video and the corresponding main tag are integrated from the collected data. For each main tag, the sum of the play time of all videos belonging to the tag is calculated.

[0055] A first media resource tag distribution is determined according to the media resource tags and the short video list.

[0056] A first probability that the visitor is a child is determined according to the first media resource tag distribution and the tag time distribution.

[0057] In this embodiment, the first probability is obtained based on a multi-layer perceptron, the input of the multi-layer perceptron is the label distribution and label time distribution of the first media resource, and the output of the multi-layer perceptron is the probability that the user is a child.

[0058] It should be noted that, in this embodiment, the first media resource is oriented towards children, and the second media resource is oriented towards adults.

[0059] Specifically, the data of the label distribution and label time distribution of the first media resource are formatted into input suitable for the machine learning model, and normalization is performed to ensure that all input features have the same scale to facilitate model processing.

[0060] The first probability is obtained based on a multi-layer perceptron model, which includes an input layer, a plurality of hidden layers and an output layer; wherein the input layer receives label distribution and time distribution data, the hidden layer is used to extract features and perform nonlinear transformation, and the output layer generates a value between 0 and 1, indicating the probability that the user is a child.

[0061] The multilayer perceptron model is trained using a labeled dataset. The multilayer perceptron model is trained using a back propagation algorithm by adjusting model parameters. The dataset includes viewing time data of media resource label distribution collected from child users and adult users.

[0062] Test the performance of the model on an independent validation set, evaluate its accuracy, recall, and F1 score, and ensure that the model is both accurate and reliable in identifying child users. Adjust the model structure or parameters based on the test results, such as adding hidden layers, changing activation functions, etc., to improve prediction performance. Deploy the trained model to the video recommendation system, receive the label time distribution data of new users in real time, and predict and update the probability of whether the user is a child.

[0063] In response to the first probability exceeding the first threshold, the video resource list in the video push list is updated to a second short video list, where the second short video list has a second media resource tag distribution.

[0064] Specifically, according to user needs, a first threshold is determined to determine whether the predicted probability is high enough to identify the user as a child. When the probability of the user being a child exceeds the first threshold, the system will filter out videos that are more suitable for adults from the video library. The selection of these videos is based on the second media resource tag distribution.

[0065] In this embodiment, the second short video list is obtained according to the following method:

[0066] The first tag list is determined from low to high according to the number of play times, wherein the number of play times refers to the number of play times of all videos under the tag; the first tag list is selected from the first category of media resource tags.

[0067] Remove duplicate tags of media resource tags corresponding to each video in the first tag list and the short video list that has been played, and obtain a second tag list;

[0068] Acquire, from a database according to the second tag list, a first short video set of short videos corresponding to the second tag list;

[0069] Obtaining a second short video set from a database according to tag information of a number of short videos played last in the first short video list, wherein the number of short videos in the second short video set is less than the number of short videos in the first short video set;

[0070] A second short video list is generated based on the first short video set and the second short video set, wherein in the second short video list, the short videos included in the second short video set are rendered to an invisible area of ​​the video recommendation component.

[0071] Specifically, the system first counts the number of views under all video tags, paying special attention to tags with low view counts, as this may reflect a smaller range of adult-oriented and professional content.

[0072] From the statistics of the number of plays, we select the first category of media resource tags that are mainly aimed at children. These tags usually contain content such as education and children's entertainment.

[0073] By comparing the list of short videos that have been played, the media resource tags that are repeated in the first tag list are removed to avoid repeated recommendations of the same type of content. At the same time, the video content is refreshed to introduce new educational or entertainment content that has not been watched in large quantities.

[0074] According to the newly sorted second tag list, the corresponding videos are extracted from the database to form the first short video set. These videos are new content for adults, not covered by the previous list, and are generally professional in nature, such as mathematics and science videos. Based on the tag information of several videos played by the user last, new video resources are extracted from the database again to form the second short video set. The number of videos in this set should be less than the first short video set.

[0075] The first and second short video sets are integrated into a new second short video list. The second short video list includes second media resource tag content related to adult users.

[0076] In order to ensure the diversity of platform content and meet the needs of audiences of different age groups, the second media resource tag content related to adult users is rendered in the invisible area of ​​the video recommendation component.

[0077] Based on the user's operation on the videos in the second short video list, updating the video list;

[0078] The updating of the video list includes updating the video resource list to a third short video list, wherein the third short video list has a third media resource tag distribution or a fourth media resource tag distribution;

[0079] In the fourth media resource tag distribution, the proportion of the first type of media resource tag distribution is higher than the proportion of the second type of media resource tag distribution;

[0080] In the second media resource tag distribution and the third media resource tag distribution, the proportion of the first type of media resource tag distribution is lower than the proportion of the second type of media resource tag distribution.

[0081] In response to the user's operation of returning to the main interface of the short video APP, a short video list is loaded on the main interface, and the short video list has a second media resource tag distribution.

[0082] According to one embodiment of the present invention, updating the video list includes the following steps:

[0083] In response to the selected duration of a video in the second short video list of the user exceeding the first duration threshold or the short video selected by the user being a member of the first short video set, randomly selecting a video from the second short video list and loading resources and acquiring a third short video list, wherein the third short video list has a third media resource tag distribution; or

[0084] In response to a selected duration of a video in the second short video list selected by the user being lower than the first duration threshold, and the short video selected by the user being a member of the first short video set, obtaining a third short video list, wherein the third short video list has a fourth media resource tag distribution; or

[0085] In response to a selected duration of a video in the second short video list selected by the user being lower than a first duration threshold, and the short video selected by the user being a member of the second short video set, a third short video list is obtained, wherein the third short video list has a third media resource tag distribution.

[0086] Specifically, the system continuously collects and analyzes the user's interactive data on the videos in the second short video list, such as viewing time, click-through rate, and number of replays. These data reflect the user's interest in and acceptance of the current recommended content. Using data analysis tools or machine learning algorithms, the user's preferred video types are identified, especially the preference for children's content and adult content.

[0087] The corresponding push content is changed based on the user's different operations on the updated second short video list. Because adults will consider many factors and have fewer records of choosing familiar similar videos, whether the user is a child is determined based on whether the user selects the videos included in the second short video list.

[0088] Specifically, if the user is slow to select and selects the second short video set, it is determined that the user is an adult. At this time, the recommendation list can be set to have a larger proportion of videos for adults or a larger proportion of videos for children, because adults can independently control the tendency of playback; there may be misjudgment in this link, but by repeatedly executing the video recommendation method of the present invention, the situation of misrecommendation can be reduced as much as possible;

[0089] If the user is slow to choose and chooses the first short video collection, they can generally be regarded as children. After analyzing the user's data, it is impossible to judge the user type based on this, and more adult videos are recommended. Even in most cases, when the user is not familiar with short video software, the recommended content can be mainly professional content. By making professional videos account for a large proportion, children will stop using short video software, while adult users will hardly feel this.

[0090] When the user selects quickly and selects the first short video set, it is determined that the user is an adult. At this time, videos for children should be added to the recommendation list. Because the video viewers are adults at this time, children mostly select familiar videos by sliding and browsing. This method can improve the user experience;

[0091] When the user selects fast and chooses the second short video collection, the user is judged to be a child. At this time, adult videos should be added to the recommendation list. By having professional videos account for a larger proportion, children will stop using short video software.

[0092] The video resource list is updated to generate a third short video list; the third short video list is determined according to the third media resource tag distribution or the fourth media resource tag distribution, including the following two situations:

[0093] In the third media resource tag, based on the user's high interest in the second media resource tag, i.e., adult-oriented, videos with these tags are selected to be added. At this time, the third short video list will contain more videos with the second media resource tag, i.e., adult-oriented;

[0094] In the fourth media resource tag, based on the user's low interest in the second media resource tag, i.e., adult-oriented, fewer videos with these tags are selected. At this time, the third short video list will contain more videos with the first media resource tag, i.e., children-oriented, but the proportion is lower than that of videos with the second media resource tag.

[0095] The system automatically adjusts the recommendation algorithm based on real-time user feedback and decides to display any of the third short video lists. By segmenting tags and monitoring user interactions, content classification is optimized to ensure that each list accurately reflects the user's latest preferences.

[0096] According to one embodiment of the present invention, an electronic device includes a processor and a memory; the memory is used to store programs; the processor executes the program to implement a video recommendation method for preventing and controlling children's addiction to short videos.

[0097] According to one embodiment of the present invention, a computer-readable storage medium stores a program, and the program is executed by a processor to implement a video recommendation method for preventing and controlling children's addiction to short videos.

[0098] The beneficial effects of the present invention are:

[0099] By precisely controlling the types of content that children can access, the platform increases the protection of child users and prevents them from being exposed to inappropriate video content;

[0100] By utilizing the user's viewing history and interaction data to dynamically update the video recommendation list, the user's personalized needs can be met more accurately.

[0101] By analyzing the distribution of label time and media resource labels of short videos watched by users, and using a multi-layer perceptron to calculate the probability of child users, the system can accurately determine whether the visitor is a child. When the recognition probability of a child user exceeds the set threshold, the system automatically updates the video push list to limit children's exposure to video content that is not suitable for their age. This intelligent recognition method improves the system's accuracy in preventing children from becoming addicted to short videos and can effectively reduce the impact of inappropriate content on children.

[0102] Based on the real-time analysis of the time distribution of short video tags and the distribution of media resource tags, the system adjusts the tag types of the video push list to achieve dynamic updates of the content. In particular, by controlling the distribution ratio of the first and second types of media resource tags, the system can gradually reduce the types of tags that are not good for children in content recommendations, while increasing the proportion of content that is more suitable for children to watch. This content tag distribution adjustment mechanism can help child users stay away from inappropriate or highly addictive content, further reducing the risk of addiction.

[0103] The system not only adjusts video recommendations based on the user's viewing behavior, but also dynamically updates the video list when the user operates (such as returning to the main interface or video selection behavior). By introducing different tag distribution mechanisms, the system can provide personalized video recommendations for children, and the content and order of recommended videos will be adjusted in real time based on user behavior. This adaptive recommendation mechanism improves the user experience and ensures that video content that complies with the anti-addiction strategy can still be pushed even when the user's interests gradually change.

[0104] The multi-layer perceptron in this technical solution outputs the probability of a child user by inputting data such as video tag distribution and viewing time. This machine learning-based model can accurately analyze massive amounts of data, greatly improving the accuracy of identifying child users. Compared with simple rules or conditional judgments, machine learning models are more efficient in processing complex behavior patterns, making the system more adaptable and intelligent in anti-addiction scenarios.

[0105] When generating the second short video list, the system removes content that is duplicated with the tags of the video that has already been played. This mechanism ensures that the pushed videos are more diverse, avoids the interference of duplicate content, and helps reduce users' over-reliance on certain types of videos. Especially for children, this method can effectively guide them away from highly addictive video content, while providing richer and more educational content choices.

[0106] Some videos in the second short video collection are rendered into invisible areas, so that the system does not push certain specific videos excessively on the surface, but dynamically displays content based on the user's operating habits. This rendering mechanism optimizes the display effect of video recommendations, while to a certain extent reducing the possibility of users watching a certain type of video for a long time, further supporting the implementation of anti-addiction functions.

[0107] Those of ordinary skill in the art will appreciate that the modules and algorithm steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0108] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and equipment can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0109] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0110] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0111] In addition, each functional module in the embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0112] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the energy-saving signal sending / receiving method of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

[0113] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the present application.

[0114] It should be understood that the size of the sequence number of each step in the content of the invention and the embodiments of the present invention does not absolutely mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention. For the purpose of example and description, the foregoing description of the implementation of the present disclosure has been given. The foregoing description is not exhaustive and is not intended to limit the present disclosure to the exact form disclosed. Various deformations and modifications may exist according to the above teachings, or various deformations and modifications may be obtained from the practice of the present disclosure. These embodiments are selected and described to illustrate the principles of the present disclosure and its practical application, so that those skilled in the art can use the present disclosure in various embodiments and various modifications suitable for the specific purpose conceived.

Claims

1. A video recommendation method for preventing and controlling children's addiction to short videos, characterized in that: include: Determine a first short video list that has been played within a first time window; Obtain the media resource tag corresponding to each video in the short video list that has been played, wherein the media resource tag includes a main tag and a sub-tag; Determine the label time distribution according to the media resource labels corresponding to the short video list and the playing time of the short videos; Determine a first media resource tag distribution according to the media resource tags and the short video list; Determine a first probability that the visitor is a child according to the first media resource tag distribution and the tag time distribution; In response to the first probability exceeding the first threshold, updating the video resource list in the video push list to a second short video list, wherein the second short video list has a second media resource tag distribution; Based on the user's operation on the videos in the second short video list, updating the video list; The updating of the video list includes updating the video resource list to a third short video list, wherein the third short video list has a third media resource tag distribution or a fourth media resource tag distribution; In the fourth media resource tag distribution, the proportion of the first type of media resource tag distribution is higher than the proportion of the second type of media resource tag distribution; In the second media resource label distribution and the third media resource label distribution, the proportion of the first type of media resource label distribution is lower than the proportion of the second type of media resource label distribution; The first probability is obtained based on a multi-layer perceptron, the input of the multi-layer perceptron is the first media resource label distribution and the label time distribution, and the output of the multi-layer perceptron is the probability that the user is a child; The second short video list is obtained in the following manner: Determine a first tag list according to the number of plays from low to high, where the number of plays is the number of plays of all videos under the tag; Remove duplicate tags of media resource tags corresponding to each video in the first tag list and the short video list that has been played, and obtain a second tag list; Acquire, from a database according to the second tag list, a first short video set of short videos corresponding to the second tag list; Obtaining a second short video set from a database according to tag information of a number of short videos played last in the first short video list, wherein the number of short videos in the second short video set is less than the number of short videos in the first short video set; Generate a second short video list based on the first short video set and the second short video set; In response to the user's operation of returning to the main interface of the short video APP, a short video list is loaded on the main interface, wherein the short video list has a second media resource tag distribution; The updating of the video list comprises the following steps: In response to the selected duration of a video in the second short video list of the user exceeding the first duration threshold or the short video selected by the user being a member of the first short video set, randomly selecting a video from the second short video list and loading resources and acquiring a third short video list, wherein the third short video list has a third media resource tag distribution; or In response to a selected duration of a video in the second short video list selected by the user being lower than the first duration threshold, and the short video selected by the user being a member of the first short video set, obtaining a third short video list, wherein the third short video list has a fourth media resource tag distribution; or In response to a selected duration of a video in the second short video list selected by the user being lower than a first duration threshold, and the short video selected by the user being a member of the second short video set, a third short video list is obtained, wherein the third short video list has a third media resource tag distribution.

2. A video recommendation method for preventing and controlling children's short video addiction as claimed in claim 1, characterized in that: The tag time distribution is determined according to the main tag corresponding to each video and the playing time of the short video.

3. A video recommendation method for preventing and controlling children's short video addiction as claimed in claim 1, characterized in that: The first tag list is selected from a first category of media resource tags.

4. A video recommendation method for preventing and controlling children's short video addiction as claimed in claim 1, characterized in that: In the second short video list, the short videos included in the second short video set are rendered to the invisible area of ​​the video recommendation component.

5. An electronic device, characterized in that: It includes a processor and a memory; the memory is used to store a program; the processor executes the program to implement the video recommendation method for preventing and controlling children's short video addiction as described in any one of claims 1-4.

6. A computer-readable storage medium, characterized in that: The storage medium stores a program, and the program is executed by a processor to implement the video recommendation method for preventing and controlling children's short video addiction as described in any one of claims 1 to 4.

Images