Information recommendation method and device, information display method and device, computer device and storage medium

Abstract

This application relates to an information recommendation method, apparatus, computer device, storage medium, and computer program product. The method includes: obtaining an information recommendation request, the information recommendation request carrying selected media data tags; obtaining various recommended media data associated with the selected media data tags according to a preset association relationship between media data and tags, and determining various candidate tags from the tags associated with each recommended media data; determining candidate media datasets associated with each candidate tag from the various recommended media data based on the preset association relationship; calculating tag recommendation scores based on the various recommended media data and the candidate media datasets associated with each candidate tag, obtaining the tag recommendation score corresponding to each candidate tag; determining a target dynamic tag from the candidate tags based on the tag recommendation score; returning the target dynamic tag to the terminal corresponding to the information recommendation request and displaying the target dynamic tag. This method can improve the accuracy of tags.

Description

Technical Field

[0001] This application relates to the field of Internet technology, and in particular to an information recommendation and information display method, device, computer equipment, storage medium, and computer program product. Background Technology

[0002] With the development of online media technology, media data search technology has emerged, enabling searchers to find the media data they need. Currently, when conducting media data searches, searchers typically receive tags for the media data, allowing them to quickly select tags and perform searches. Currently, fixed tags are commonly used for media data searches. However, due to the rapid changes in data in the internet age, prolonged use of fixed tags can lead to a decrease in tag accuracy, thereby reducing search accuracy. Summary of the Invention

[0003] Therefore, it is necessary to provide an information recommendation and display method, device, computer equipment, computer-readable storage medium, and computer program product that can improve the accuracy of labels in order to address the above-mentioned technical problems.

[0004] On the one hand, this application provides an information recommendation method. The method includes:

[0005] Obtain an information recommendation request, which includes the selected media data tags;

[0006] According to the preset association relationship between media data and tags, obtain each recommended media data associated with the selected media data tags, and determine each candidate tag from the tags associated with each recommended media data. The tags are used to characterize the attribute information of the media data.

[0007] Based on preset association relationships, candidate media datasets associated with each candidate tag are determined from various recommended media data.

[0008] The tag recommendation score is calculated based on the candidate media dataset associated with each recommended media data and each candidate tag, and the tag recommendation score corresponding to each candidate tag is obtained.

[0009] Based on the tag recommendation score, the target dynamic tag is determined from each candidate tag, and the target dynamic tag is returned to the terminal corresponding to the information recommendation request so that the target dynamic tag can be displayed on the terminal.

[0010] In one embodiment, the information recommendation method further includes:

[0011] When all initial candidate tags are preset static display tags, a dynamic tag not found reminder message is generated and returned to the terminal so that the reminder message can be displayed on the terminal.

[0012] In one embodiment, obtaining each pair of similar tags includes:

[0013] Retrieve each set tag, pair each set tag together to obtain tag pairs;

[0014] Calculate the point-to-point mutual information for each label pair, and label pairs whose point-to-point mutual information exceeds a preset information threshold are considered similar label pairs.

[0015] In one embodiment, each candidate tag is determined from the tags associated with each recommended media data, including:

[0016] Obtain the interest tags corresponding to the target object, and determine each candidate tag based on the interest tags and the tags associated with each recommended media data.

[0017] On the other hand, this application also provides an information recommendation device. The device includes:

[0018] The request retrieval module is used to retrieve information recommendation requests, which carry the selected media data tags.

[0019] The candidate tag determination module is used to obtain each recommended media data associated with the selected media data tags according to the preset association relationship between media data and tags, and to determine each candidate tag from the tags associated with each recommended media data. The tags are used to characterize the attribute information of the media data.

[0020] The data determination module is used to determine the candidate media dataset associated with each candidate tag from each recommended media data based on preset association relationships;

[0021] The recommendation score calculation module is used to calculate the tag recommendation score based on each recommended media data and the candidate media dataset associated with each candidate tag, and obtain the tag recommendation score corresponding to each candidate tag.

[0022] The tag recommendation module is used to determine the target dynamic tag from various candidate tags based on the tag recommendation degree, and return the target dynamic tag to the terminal corresponding to the information recommendation request so that the target dynamic tag can be displayed on the terminal.

[0023] On the other hand, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0024] Obtain an information recommendation request, which includes the selected media data tags;

[0025] According to the preset association relationship between media data and tags, obtain each recommended media data associated with the selected media data tags, and determine each candidate tag from the tags associated with each recommended media data. The tags are used to characterize the attribute information of the media data.

[0026] Based on preset association relationships, candidate media datasets associated with each candidate tag are determined from various recommended media data.

[0027] The tag recommendation score is calculated based on the candidate media dataset associated with each recommended media data and each candidate tag, and the tag recommendation score corresponding to each candidate tag is obtained.

[0028] Based on the tag recommendation score, the target dynamic tag is determined from each candidate tag, and the target dynamic tag is returned to the terminal corresponding to the information recommendation request so that the target dynamic tag can be displayed on the terminal.

[0029] On the other hand, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0030] Obtain an information recommendation request, which includes the selected media data tags;

[0031] According to the preset association relationship between media data and tags, obtain each recommended media data associated with the selected media data tags, and determine each candidate tag from the tags associated with each recommended media data. The tags are used to characterize the attribute information of the media data.

[0032] Based on preset association relationships, candidate media datasets associated with each candidate tag are determined from various recommended media data.

[0033] The tag recommendation score is calculated based on the candidate media dataset associated with each recommended media data and each candidate tag, and the tag recommendation score corresponding to each candidate tag is obtained.

[0034] Based on the tag recommendation score, the target dynamic tag is determined from each candidate tag, and the target dynamic tag is returned to the terminal corresponding to the information recommendation request so that the target dynamic tag can be displayed on the terminal.

[0035] On the other hand, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0036] Obtain an information recommendation request, which includes the selected media data tags;

[0037] According to the preset association relationship between media data and tags, obtain each recommended media data associated with the selected media data tags, and determine each candidate tag from the tags associated with each recommended media data. The tags are used to characterize the attribute information of the media data.

[0038] Based on preset association relationships, candidate media datasets associated with each candidate tag are determined from various recommended media data.

[0039] The tag recommendation score is calculated based on the candidate media dataset associated with each recommended media data and each candidate tag, and the tag recommendation score corresponding to each candidate tag is obtained.

[0040] Based on the tag recommendation score, the target dynamic tag is determined from each candidate tag, and the target dynamic tag is returned to the terminal corresponding to the information recommendation request so that the target dynamic tag can be displayed on the terminal.

[0041] The aforementioned information recommendation method, apparatus, computer equipment, storage medium, and computer program product, by acquiring an information recommendation request carrying selected media data tags, obtaining recommended media data associated with the selected media data tags according to a preset association relationship between media data and tags, and determining candidate tags from the tags associated with each recommended media data (tags are used to characterize the attribute information of the media data); determining candidate media datasets associated with each candidate tag from each recommended media data based on the preset association relationship; calculating tag recommendation scores based on each recommended media data and each candidate tag's associated candidate media dataset; determining target dynamic tags from each candidate tag based on the tag recommendation scores; and returning the target dynamic tags to the terminal corresponding to the information recommendation request for display on the terminal. By determining each candidate tag through selected media data tags, calculating the recommendation score of the candidate tags, determining the target dynamic tag based on the recommendation score, and then displaying the target dynamic tag on the terminal, the accuracy of tags is improved, the use of fixed tags is avoided, and dynamic tags can be used for media data searching, thereby improving the accuracy of media data searching.

[0042] On the one hand, this application provides an information display method. The method includes:

[0043] Display various static display tags on the media data search page;

[0044] In response to the selection event of the target static display tag among the various static display tags, the target dynamic tag is displayed on the media data search page. The target dynamic tag is determined by identifying each candidate tag from the tags corresponding to each recommended media data associated with the target static display tag, calculating the recommendation degree based on each recommended media data associated with the target static display tag and the candidate media datasets associated with each candidate tag, and determining the tag recommendation degree from each candidate tag based on the tag recommendation degree. The tag is used to represent the attribute information of the media data.

[0045] On the other hand, this application also provides an information display device. The device includes:

[0046] The static tag display module is used to display various static tags on the media data search page;

[0047] The dynamic tag display module is used to display the target dynamic tag on the media data search page in response to the selection event of the target static display tag among the various static display tags. The target dynamic tag is determined by identifying each candidate tag from the tags corresponding to each recommended media data associated with the target static display tag, calculating the recommendation degree based on each recommended media data associated with the target static display tag and the candidate media datasets associated with each candidate tag, and determining the tag recommendation degree from each candidate tag based on the tag recommendation degree. The tag is used to represent the attribute information of the media data.

[0048] On the other hand, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0049] Display various static display tags on the media data search page;

[0050] In response to the selection event of the target static display tag among the various static display tags, the target dynamic tag is displayed on the media data search page. The target dynamic tag is determined by identifying each candidate tag from the tags corresponding to each recommended media data associated with the target static display tag, calculating the recommendation degree based on each recommended media data associated with the target static display tag and the candidate media datasets associated with each candidate tag, and determining the tag recommendation degree from each candidate tag based on the tag recommendation degree. The tag is used to represent the attribute information of the media data.

[0051] On the other hand, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0052] Display various static display tags on the media data search page;

[0053] In response to the selection event of the target static display tag among the various static display tags, the target dynamic tag is displayed on the media data search page. The target dynamic tag is determined by identifying each candidate tag from the tags corresponding to each recommended media data associated with the target static display tag, calculating the recommendation degree based on each recommended media data associated with the target static display tag and the candidate media datasets associated with each candidate tag, and determining the tag recommendation degree from each candidate tag based on the tag recommendation degree. The tag is used to represent the attribute information of the media data.

[0054] On the other hand, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0055] Display various static display tags on the media data search page;

[0056] In response to the selection event of the target static display tag among the various static display tags, the target dynamic tag is displayed on the media data search page. The target dynamic tag is determined by identifying each candidate tag from the tags corresponding to each recommended media data associated with the target static display tag, calculating the recommendation degree based on each recommended media data associated with the target static display tag and the candidate media datasets associated with each candidate tag, and determining the tag recommendation degree from each candidate tag based on the tag recommendation degree. The tag is used to represent the attribute information of the media data.

[0057] The aforementioned information display method, apparatus, computer equipment, storage medium, and computer program product display various static display tags on a media data search page; in response to a selection event of a target static display tag among the various static display tags, a target dynamic tag is displayed on the media data search page; wherein, the target dynamic tag is determined by identifying candidate tags from the tags corresponding to the various recommended media data associated with the target static display tag, calculating the recommendation degree based on the recommendation degree of each candidate media dataset associated with the target static display tag and the candidate media dataset associated with each candidate tag respectively, and determining the tag recommendation degree from each candidate tag based on the tag recommendation degree. The tag is used to characterize the attribute information of the media data. By determining each candidate tag when the target static display tag is selected, then calculating the recommendation degree of the candidate tags, determining the target dynamic tag based on the recommendation degree, and then displaying the target dynamic tag, the accuracy of the tag is improved, avoiding only displaying fixed tags. Then, dynamic tags can be used for media data search, thereby improving the accuracy of media data search. Attached Figure Description

[0058] Figure 1 This is a diagram illustrating the application environment of an information recommendation method in one embodiment.

[0059] Figure 2 This is a flowchart illustrating an information recommendation method in one embodiment;

[0060] Figure 3 This is a schematic diagram of the process for obtaining each candidate label in one embodiment;

[0061] Figure 4 This is a schematic diagram of the process for obtaining tag recommendation scores in one embodiment;

[0062] Figure 5 This is a schematic diagram of the process for obtaining various dissimilar labels in one embodiment;

[0063] Figure 6 This is a flowchart illustrating the media data search page in one embodiment;

[0064] Figure 7 This is a flowchart illustrating the process of recommending media data for a target in one embodiment;

[0065] Figure 8 This is a flowchart illustrating an information display method in one embodiment;

[0066] Figure 9 This is a schematic diagram of a media data search page obtained in a specific embodiment;

[0067] Figure 10 This is a schematic diagram of a media data search page obtained in another specific embodiment;

[0068] Figure 11 This is a schematic diagram of a media data search page obtained in yet another specific embodiment;

[0069] Figure 12 This is a schematic diagram of a media data search page obtained in another specific embodiment;

[0070] Figure 13 This is a flowchart illustrating an information recommendation method in a specific embodiment;

[0071] Figure 14 This is a schematic diagram of the architecture of an information recommendation method in a specific embodiment;

[0072] Figure 15 This is a structural block diagram of an information recommendation device in one embodiment;

[0073] Figure 16 This is a structural block diagram of an information display device in one embodiment;

[0074] Figure 17 This is an internal structural diagram of a computer device in one embodiment;

[0075] Figure 18 This is a diagram of the internal structure of a computer device in another embodiment. Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0077] The information recommendation method provided in this application embodiment can be applied to, for example, Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on the cloud or other servers. Server 104 receives an information recommendation request sent by terminal 102, which carries selected media data tags. Server 104 retrieves each recommended media data associated with the selected media data tags according to the preset association relationship between media data and tags, and determines each candidate tag from the tags associated with each recommended media data. Tags are used to characterize the attribute information of the media data. Server 104 determines the candidate media datasets associated with each candidate tag from each recommended media data based on the preset association relationship. Server 104 calculates the tag recommendation degree based on each recommended media data and the candidate media datasets associated with each candidate tag, obtaining the tag recommendation degree corresponding to each candidate tag. Server 104 determines the target dynamic tag from each candidate tag based on the tag recommendation degree and returns the target dynamic tag to the terminal corresponding to the information recommendation request, so that the target dynamic tag can be displayed on the terminal. The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle systems. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0078] In one embodiment, such as Figure 2 As shown, an information recommendation method is provided, which can be applied to... Figure 1 Taking a server as an example, it can be understood that this method can also be applied to a terminal, and also to a system that includes both a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0079] Step 202: Obtain the information recommendation request, which carries the selected media data tags.

[0080] The information recommendation request is used for dynamic tag recommendations. Selected media data tags refer to the tags corresponding to the selected media data. These tags can include only static display tags, only dynamic tags, or both. The static display tags are pre-displayed to the requester's terminal and are pre-set tags that will not change during display on the terminal. The dynamic tags change during display. The media data refers to the medium through which information is disseminated via an internet information dissemination platform. This media data can be image data, text data, video data, audio data, etc.

[0081] Specifically, the terminal can display pre-set static display tags on the media data search page. These static display tags can include multiple tags. When the terminal detects that a searcher has selected a displayed static display tag, it sends an information recommendation request to the server. This information recommendation request carries the selected static display tag, i.e., the selected media data tag. The server receives the information recommendation request, parses it, and obtains the selected media data tag. In one embodiment, the searcher can also select multiple static display tags through the terminal. The server then obtains all selected static display tags carried in the information recommendation request. In one embodiment, when dynamic tags are displayed on the terminal, only the dynamic tag can be selected, and the terminal uses the selected dynamic tag as the selected media data tag. In one embodiment, when both static and dynamic tags are displayed on the terminal, both static and dynamic tags can be selected simultaneously, and the terminal uses both static and dynamic tags as the selected media data tags.

[0082] Step 204: Obtain each recommended media data associated with the selected media data tags according to the preset association relationship between media data and tags, and determine each candidate tag from the tags associated with each recommended media data. The tags are used to characterize the attribute information of the media data.

[0083] The pre-defined association between media data and tags refers to establishing a correspondence between media data and tags in advance. Media data can be associated with multiple different tags, each tag representing different attribute information of the media data. Different media data can have the same tag or different tags. The tag represents the attribute information of the media data, which refers to the information represented by the specific content of the media data. For example, a tag can be a tag for the category to which the media data belongs, or a tag for the content described by the media data. Candidate tags refer to tags that need further filtering; these candidate tags are different from the statically displayed tags.

[0084] Specifically, the server retrieves recommended media data associated with the selected media data tags from the media database according to the preset association between media data and tags. This means that the tags associated with each recommended media data include the selected media data tags. Then, the server determines all tags associated with each recommended media data according to the preset association between media data and tags, and finally selects candidate tags from all the tags associated with each recommended media data.

[0085] Step 206: Based on the preset association relationship, determine the candidate media datasets associated with each candidate tag from each recommended media data.

[0086] The candidate media dataset includes various candidate media data, which refers to the media data associated with the candidate tags.

[0087] Specifically, the server filters from various recommended media data according to preset association relationships. When the tags associated with recommended media data include candidate tags, the recommended media data is used as the candidate media data corresponding to that candidate tag. The server filters from each recommended media data to obtain all candidate media data corresponding to each candidate tag, resulting in a candidate media dataset associated with each candidate tag. This candidate media dataset may contain only one candidate media data.

[0088] Step 208: Calculate the tag recommendation score based on the candidate media dataset associated with each recommended media data and each candidate tag, and obtain the tag recommendation score corresponding to each candidate tag.

[0089] The tag recommendation score represents the likelihood that a candidate tag will be recommended to the terminal for display. The higher the tag recommendation score, the greater the likelihood that the candidate tag will be recommended to the terminal for display. Different candidate tags correspond to different candidate media datasets, resulting in different tag recommendation scores.

[0090] Specifically, the server calculates the data quality of the recommended media data using each piece of recommended media data. Then, it calculates the data quality of the candidate media dataset based on the candidate media dataset associated with each candidate tag. Finally, it calculates the ratio of the data quality of the candidate media dataset to the data quality of the recommended media data to obtain the tag recommendation score for each candidate tag. Data quality can be represented by the quantity of media data or by the content quality score of the media data.

[0091] Step 210: Determine the target dynamic tag from each candidate tag based on the tag recommendation degree, and return the target dynamic tag to the terminal corresponding to the information recommendation request so that the target dynamic tag can be displayed on the terminal.

[0092] The target dynamic tag refers to the dynamic tag to be displayed to the terminal. A dynamic tag is a tag that changes when displayed on the terminal. The target dynamic tag determined based on the static display tag will also be different if the static display tag is carried in the request.

[0093] Specifically, the server can select candidate tags with the highest recommendation scores as target dynamic tags based on their recommendation scores. These target dynamic tags are then returned to the terminal corresponding to the recommendation request for display. The server can then retrieve associated target recommended media data from various recommended media data sources based on the selected dynamic tags. This data can be returned to the terminal, which can then display it, thereby improving the accuracy of media data recommendations.

[0094] In one embodiment, the server can sequentially determine at least two target dynamic tags from each candidate tag based on the tag recommendation degree, and return the at least two target dynamic tags to the terminal corresponding to the information recommendation request, so as to display the at least two target dynamic tags through the terminal, thereby improving the accuracy of tag recommendation and enabling searchers to flexibly select target dynamic tags.

[0095] In one embodiment, the information recommendation request is also used to request the recommended media data associated with the selected media data tag. After obtaining the target dynamic tag, the server can return the target dynamic tag and the recommended media data to the terminal corresponding to the information recommendation request, so that the target dynamic tag and the recommended media data can be displayed on the terminal. At the same time, the server can return the recommended media data associated with the selected media data tag, which can further facilitate the searcher's use and improve the experience.

[0096] The aforementioned information recommendation method, apparatus, computer equipment, storage medium, and computer program product, by acquiring an information recommendation request carrying selected media data tags, obtaining recommended media data associated with the selected media data tags according to a preset association relationship between media data and tags, and determining candidate tags from the tags associated with each recommended media data (tags are used to characterize the attribute information of the media data); determining candidate media datasets associated with each candidate tag from each recommended media data based on the preset association relationship; calculating tag recommendation scores based on each recommended media data and each candidate tag's associated candidate media dataset; determining target dynamic tags from each candidate tag based on the tag recommendation scores; and returning the target dynamic tags to the terminal corresponding to the information recommendation request for display on the terminal. By determining each candidate tag through selected media data tags, calculating the recommendation score of the candidate tags, determining the target dynamic tag based on the recommendation score, and then displaying the target dynamic tag on the terminal, the accuracy of tags is improved, the use of fixed tags is avoided, and dynamic tags can be used for media data searching, thereby improving the accuracy of media data searching.

[0097] In one embodiment, such as Figure 3 As shown, step 204 involves determining candidate tags from the tags associated with each recommended media data, including:

[0098] Step 302: Obtain the tags associated with each recommended media data according to the preset association relationship, filter the selected media data tags from the tags associated with each recommended media data, and obtain each initial candidate tag.

[0099] Specifically, the server can retrieve the tags associated with each recommended media data from the database according to the relationship between media data and tags. Different recommended media data are associated with different tags, and each recommended media data is associated with multiple different tags. Then, the selected media data tags are removed from the tags associated with each recommended media data, and the remaining tags are used as the initial candidate tags.

[0100] Step 304: When there is no preset static display tag among the initial candidate tags, each initial candidate tag is used as a candidate tag.

[0101] Step 306: When there are preset static display tags among the initial candidate tags, filter the preset static display tags among the initial candidate tags to obtain each candidate tag.

[0102] Among them, preset static display tags refer to pre-set static display tags, which include multiple tags. Media data tags can be pre-set into two categories: static display tags and dynamic tags. Tags associated with media data can include both static and dynamic tags.

[0103] Specifically, the server determines whether each initial candidate tag has a static display tag. If no preset static display tag exists for any initial candidate tag, it means that none of the initial candidate tags are displayed on the terminal. In this case, all initial candidate tags are used as candidate tags. If a preset static display tag exists for any initial candidate tag, it means that some of the initial candidate tags have already been displayed on the terminal. In this case, the preset static display tags are filtered out, and the remaining initial candidate tags are used as candidate tags. In a specific embodiment, the media data is video, and the selected media data tags include historical drama tags and romance tags. At this time, the historical drama tag and the romance tag obtain the associated video a and video b. Then, video a also contains a martial arts tag, and video b also contains a fantasy tag. When neither the martial arts tag nor the fantasy tag is a static tag, the martial arts tag and the fantasy tag are used as candidate tags.

[0104] In the above embodiments, candidate tags are determined by judging whether there are preset static display tags among each initial candidate tag, thereby ensuring that the final determined target dynamic tag is different from the displayed static display tag, thus improving the accuracy of the target dynamic tag.

[0105] In one embodiment, the information recommendation method further includes:

[0106] When all initial candidate tags are preset static display tags, a dynamic tag not found reminder message is generated and returned to the terminal so that the reminder message can be displayed on the terminal.

[0107] The "Dynamic Tag Not Found" reminder message refers to a reminder message used to indicate that the selected media data tag does not have a corresponding dynamic tag. This message can be a reminder text, a reminder voice, a reminder image, or blank information.

[0108] Specifically, the server determines that each initial candidate tag can be found in the pre-set static display tags. If so, it means that all the determined initial candidate tags are static display tags. The server then directly generates a "dynamic tag not found" notification and returns this notification to the terminal for display. This notification reminds the searcher to reselect a static display tag.

[0109] In one embodiment, when the recommended media data associated with the selected media data tag only contains the selected media data tag, a dynamic tag not found reminder message can also be generated and returned to the terminal for display.

[0110] In the above embodiments, when each initial candidate tag is a preset static display tag, a dynamic tag not found reminder message is generated and returned to the terminal so that the dynamic tag not found reminder message can be displayed on the terminal, which facilitates timely re-search selection and improves search efficiency.

[0111] In one embodiment, step 206, determining the candidate media dataset associated with each candidate tag from each recommended media dataset based on a preset association relationship, includes the following steps:

[0112] The current candidate label is determined from the candidate labels. Recommended media data associated with the current candidate label are filtered from the recommended media data according to the preset association relationship to obtain the candidate media dataset associated with the current candidate label. The candidate labels are traversed to obtain the candidate media dataset associated with each candidate label.

[0113] Among them, the current candidate label refers to the candidate label for the currently identified associated media data.

[0114] Specifically, the server sequentially filters candidate media data for each candidate tag. The server can first determine the current candidate tag from all candidate tags, and then filter the recommended media data associated with the current candidate tag from all recommended media data according to preset association relationships to obtain the candidate media dataset associated with the current candidate tag. Then, the server determines the current candidate tag from the remaining candidate tags and filters the associated recommended media data until all candidate tags have been filtered to obtain the associated candidate media dataset.

[0115] In the above embodiments, by filtering recommended media data associated with the current candidate tag from various recommended media data, and then traversing each candidate tag to obtain the candidate media dataset associated with each candidate tag, the accuracy of the obtained candidate media dataset can be improved.

[0116] In one embodiment, step 208, calculating the tag recommendation score based on each recommended media data and the candidate media dataset associated with each candidate tag, to obtain the tag recommendation score corresponding to each candidate tag, includes the following steps:

[0117] Calculate the number of recommendations corresponding to each recommended media data, and calculate the number of candidates in the candidate media dataset associated with each candidate tag; calculate the ratio of the number of candidates corresponding to each candidate tag to the number of recommendations, and obtain the tag recommendation degree corresponding to each candidate tag.

[0118] The number of recommendations refers to the number of recommended media data. The number of candidates refers to the number of candidate media data in the candidate media dataset.

[0119] Specifically, the server counts the number of data points for each recommended media item to obtain the recommendation count. Simultaneously, it counts the number of candidate media items in the candidate media dataset associated with each candidate tag to obtain the candidate count for each candidate tag. Then, it calculates the ratio of the candidate count for each candidate tag to the recommendation count, and uses this ratio as the tag recommendation score for each candidate tag.

[0120] In the above embodiments, calculating the tag recommendation score by calculating the amount of data can improve the efficiency of obtaining the tag recommendation score.

[0121] In one embodiment, such as Figure 4 As shown, step 208 involves calculating the tag recommendation score based on the candidate media dataset associated with each recommended media data and each candidate tag, to obtain the tag recommendation score corresponding to each candidate tag, including:

[0122] Step 402: Obtain the data quality information corresponding to each recommended media data, and obtain the candidate quality information corresponding to the candidate media data in the candidate media dataset associated with each candidate label.

[0123] Specifically, data quality information characterizes the quality of recommended media data. This information can be represented by a quality score, which measures the recommendability of the media data. The better the quality of the media data, the higher its recommendability. This quality information can be calculated based on the interaction information and features of the media data. Candidate quality information characterizes the quality of candidate media data.

[0124] Specifically, the server can retrieve the data quality information corresponding to each recommended media data from the database. Simultaneously, it can retrieve the candidate quality information corresponding to the candidate media data in the candidate media dataset associated with each candidate tag. After obtaining each media data, the server can extract the attribute features and interaction features corresponding to each media data, calculate the quality information corresponding to each media data based on these features, and then save the quality information corresponding to each media data. When needed, searching from the saved quality information can improve efficiency. The attribute features corresponding to the media data represent the inherent attributes of the media data itself. The interaction features corresponding to the media data represent the interaction information between the media data and the interactors. For example, when the media data is video data, the attribute features could be video quality, frame rate, duration, etc., and the interaction features could be viewing time, collection rate, forwarding rate, like rate, etc.

[0125] Step 404: Sum the data quality information corresponding to each recommended media data to obtain the total data quality information, and sum the information based on the candidate quality information corresponding to the candidate media data in the candidate media dataset to obtain the candidate total quality information corresponding to each candidate label.

[0126] The overall data quality information refers to the sum of the data quality of all recommended media data. The overall candidate quality information refers to the sum of the data quality of all candidate media data corresponding to the candidate tags.

[0127] Specifically, the server calculates the sum of the quality information for each recommended media data set to obtain the overall data quality information. Simultaneously, it calculates the sum of the quality information for each candidate media data set corresponding to each candidate tag to obtain the overall candidate quality information for that candidate tag.

[0128] Step 406: Based on the candidate total quality information and data total quality information corresponding to each candidate label, obtain the label recommendation degree corresponding to each candidate label.

[0129] Specifically, the server calculates the ratio of the total quality information of each candidate to the total quality information of the data, and obtains the tag recommendation degree corresponding to each candidate tag.

[0130] In a specific embodiment, the tag recommendation degree can be calculated using the formula (1) shown below.

[0131]

[0132] Here, "score" refers to the tag recommendation score, D1 refers to the set of candidate media data corresponding to the candidate tags, and D2 refers to the set of data for each recommended media. dThis refers to the quality score of media data. This refers to calculating the total quality of candidate media data in a candidate media dataset. This refers to the sum of the quality information of each recommendation media data point. When weight... d When the value is 1, it indicates the number of media data to be calculated.

[0133] In the above embodiments, by calculating the total candidate quality information and total data quality information corresponding to each candidate tag, and then calculating the proportion of the total candidate quality information corresponding to each candidate tag to the total data quality information, the tag recommendation degree corresponding to each candidate tag is obtained. That is, the data quality information of media data is used to calculate the tag recommendation degree, which improves the accuracy of the obtained tag recommendation degree.

[0134] In one embodiment, such as Figure 5 As shown, step 210 involves determining the target dynamic tag from among the candidate tags based on the tag recommendation score, and returning the target dynamic tag to the terminal corresponding to the information recommendation request so that the target dynamic tag can be displayed on the terminal. This includes:

[0135] Step 502: Select from each candidate tag in turn based on the tag recommendation degree to obtain at least two target dynamic tags.

[0136] Specifically, the server can sort the candidate tags according to their recommendation level, and then select the top candidate tags as target dynamic tags. At least two candidate tags can be selected to obtain at least two target dynamic tags.

[0137] Step 504: Obtain each similar tag pair, and based on each similar tag pair, filter out dissimilar tags from at least two target dynamic tags to obtain each dissimilar tag.

[0138] Step 506: Recommend the terminal corresponding to each dissimilar tag information return request, so that each dissimilar tag can be displayed on the terminal.

[0139] The similarity tag pair includes two tags whose similarity exceeds a preset similarity threshold. This similarity can be calculated by performing similarity calculations on all tags or on all dynamic tags. Dissimilar tags are not similar to any of the target dynamic tags other than themselves; this dissimilarity means that the similarity is less than the preset similarity threshold.

[0140] Specifically, the server can retrieve similar tag pairs from the database, as well as from similar tag pairs uploaded by the terminal. The server can also perform similarity calculations on the retrieved tags to obtain similar tag pairs. Then, it uses these similar tag pairs to filter for dissimilar tags from at least two target dynamic tags. Specifically, it matches similar tag pairs from at least two target dynamic tags. When searching for two tags in a similar tag pair from at least two target dynamic tags, it filters out either of those two tags until the remaining target dynamic tags are dissimilar to each other. These remaining target dynamic tags are then designated as dissimilar tags. The server then returns the dissimilar tag information to the terminal corresponding to the recommendation request, so that the dissimilar tags can be displayed on the terminal.

[0141] In the above embodiments, when at least two target dynamic tags are obtained through filtering, each similar tag pair is acquired, and then the similar dynamic tags among the at least two target dynamic tags are filtered using each similar tag pair to obtain each dissimilar tag. Then, each dissimilar tag is displayed, thereby avoiding the display of duplicate dynamic tags and improving the accuracy of dynamic tag display.

[0142] In one embodiment, step 504, obtaining each pair of similar tags, includes the following steps:

[0143] Get each set label, pair each set label to get each label pair; get the co-occurrence count of each label pair, and label pairs with co-occurrence counts exceeding a preset threshold are considered similar label pairs.

[0144] Here, "pre-defined tags" refers to the tags pre-set in the media data, including all dynamic tags and all static display tags. A tag pair is a combination of two pre-defined tags. Co-occurrence count refers to the number of times the two pre-defined tags in a tag pair appear together. Co-occurrence means that the tags associated with the same media data simultaneously include both pre-defined tags from the tag pair; the co-occurrence count is obtained by counting all media data that simultaneously include both pre-defined tags from the tag pair. The preset co-occurrence count threshold is a pre-set threshold that can be set according to requirements.

[0145] Specifically, the server retrieves all configured tags from the database, or from tags uploaded by the terminal. The server can also obtain configured tags from the data service provider. Then, the configured tags are paired, which can be done randomly. The server pairs each configured tag with all other configured tags to obtain tag pairs. The server can then retrieve the co-occurrence count for each tag pair from the database. Alternatively, the server can count the co-occurrence count for each tag pair. The co-occurrence count for each tag pair is then compared to a preset threshold. If the co-occurrence count exceeds the preset threshold, the tag pair with the higher co-occurrence count is considered a similar tag pair.

[0146] In the above embodiments, by obtaining the co-occurrence counts corresponding to each tag pair, tag pairs with co-occurrence counts exceeding a preset threshold are regarded as similar tag pairs, thereby improving the efficiency of obtaining similar tag pairs.

[0147] In one embodiment, step 504, obtaining each pair of similar tags, includes the following steps:

[0148] Obtain each set label, pair each set label to obtain each label pair; calculate the point-to-point mutual information corresponding to each label pair, and identify label pairs whose point-to-point mutual information exceeds a preset information threshold as similar label pairs.

[0149] Among them, "pre-set tags" refers to tags pre-defined in the media data, which can include dynamic tags and static display tags. Static display tags are tags that remain fixed on the terminal display; they are identical across different terminals and on the same page. Dynamic tags, on the other hand, change based on their recommendation level. When a tag's recommendation level changes, the dynamic tags displayed on the terminal also change accordingly. Point-to-point mutual information refers to the information between two tags in a tag pair calculated using a point-to-point mutual information algorithm. This algorithm measures the relevance between two tags in a tag pair and can be a PMI (Pointwise Mutual Information) algorithm. The preset information threshold is a pre-set threshold for point-to-point mutual information. When the point-to-point mutual information exceeds the preset threshold, it indicates that the two tags in the tag pair are similar; when the point-to-point mutual information does not exceed the preset threshold, it indicates that the two tags in the tag pair are dissimilar.

[0150] Specifically, the server can retrieve each set tag directly from the database, from the business service provider, from the terminal's uploaded tags, or from the data provider. Then, the server pairs these tags to obtain tag pairs. The server uses a point-to-point mutual information algorithm to calculate the point-to-point mutual information for each tag pair, and tags with mutual information exceeding a preset threshold are considered similar tag pairs.

[0151] In a specific embodiment, the mutual information between the point pairs corresponding to the tag pairs can be calculated using the following formula (2).

[0152]

[0153] Here, x refers to one label in a label pair, y refers to the other label in a label pair, p(x) represents the probability of label x appearing, p(y) represents the probability of label y appearing, and p(x,y) represents the probability of both labels x and y appearing simultaneously. By statistically analyzing the probability of two labels appearing simultaneously in the corresponding labels of the media data, the higher the probability, the higher the similarity of the label pair.

[0154] In the above embodiments, by calculating the point-to-point mutual information of the tag pairs, tag pairs whose point-to-point mutual information exceeds a preset information threshold are regarded as similar tag pairs. The point-to-point mutual information is used to determine similar tag pairs, thereby improving the accuracy of the obtained similar tag pairs.

[0155] In one embodiment, step 504, obtaining each pair of similar tags, includes the following steps:

[0156] Obtain each set label and its corresponding co-occurrence information. Build a label relationship graph based on each set label and its corresponding co-occurrence information. Select label node pairs from the label relationship graph and calculate the density of the label node pairs based on the label relationship graph. Label node pairs with a density exceeding a preset density threshold are considered similar label pairs. The density is calculated based on the connection edge weight between the two label nodes in the label node pair and the outgoing edge weight of each label node.

[0157] Co-occurrence information is used to characterize the information that set tags together serve as tags for the same media data. This can include co-occurrence relationships and co-occurrence counts. Co-occurrence relationships characterize tags that share the same media data. A tag relationship graph is a graph where set tags are nodes, co-occurrence relationships are edges connecting nodes, and co-occurrence counts are edge weights. A tag node pair refers to the tags corresponding to two nodes in the tag relationship graph. The tightness of a tag node pair characterizes the similarity between the two tags in the pair, calculated using the edge weights in the tag relationship graph. Higher tightness indicates greater similarity between the two tags in the pair. A preset tightness threshold is a pre-set threshold for tightness. If the tightness exceeds the preset threshold, the two tags in the pair are similar; conversely, if the tightness does not exceed the preset threshold, the two tags in the pair are dissimilar.

[0158] Specifically, the server retrieves all set tags from the database, and then retrieves the co-occurrence information corresponding to each set tag. It then uses the set tags and their co-occurrence information to build a tag relationship graph, where each set tag is treated as a node in the graph. The nodes are connected according to the co-occurrence relationships in the co-occurrence information to obtain edges, and the number of co-occurrences is used as the edge weights. Next, tag node pairs are selected sequentially from the tag relationship graph, and the weights of the connecting edges between the two tag nodes and the outgoing edges of each tag node are determined. The density of the tag node pairs is then calculated based on these weights. Finally, tag node pairs with a density exceeding a preset density threshold are designated as similar tag pairs.

[0159] In one specific embodiment, the tightness of the tag pair can be calculated using the formula (3) shown below.

[0160]

[0161] Where S(x,y) represents the tightness between label x and label y in a label pair. x,y D represents the weight of the edges connecting corresponding nodes in the label relationship graph. x Let D represent the set of outgoing edges of the node corresponding to label x in the label graph. y This represents the set of outgoing edges of the node corresponding to label y in the label relationship graph. When the weight of the edges connecting label nodes exceeds a preset density threshold, the label pair is considered a similar label pair.

[0162] In the above embodiments, by establishing a label relationship graph, then using the label relationship graph to calculate the density of label node pairs, and then using the density to determine similar label pairs, the accuracy of the obtained similar label pairs can be improved.

[0163] In one embodiment, such as Figure 6 The method further includes, in step 202, prior to obtaining the information recommendation request:

[0164] Step 602: Obtain the display request for the media data search page, with the display request carrying the target object.

[0165] Step 604: Obtain each static display tag based on the display request, and obtain each initial media data based on the target object and each static display tag.

[0166] The media data search page is the page for searching media data. This page displays tags used for searching media data, including static and dynamic tags. It also displays media data obtained through tag searches. The target object represents the object being searched for; it can be a virtual object or a real object. Virtual objects include virtual idols, virtual broadcasters, etc. Real objects can be, for example, people. Initial media data refers to the media data displayed when the media data search page is initialized.

[0167] Specifically, when the server receives a media data search page display request sent by the target object through the terminal, it retrieves various static display tags from the database based on the display request, and then obtains various initial media data based on the target object and each static display tag. These initial media data can be media data associated with each static display tag, sorted according to the target object's level of interest. Alternatively, the initial media data can be sorted according to the popularity of the media data and displayed on the terminal.

[0168] Step 606: Return each initial media data and each static display tag to the terminal corresponding to the display request, so that the terminal can display each static display tag and each initial media data on the media data search page.

[0169] Specifically, the server responds to the display request by returning each initial media data and each static display tag to the terminal corresponding to the display request. The terminal receives the initial media data and each static display tag returned by the server and displays each static display tag and each initial media data on the media data search page. Specifically, each static display tag can be displayed in the tag display area of ​​the media data search page, and each initial media data can be displayed in the media data display area of ​​the media data search page.

[0170] In one embodiment, the server can also obtain pre-set personalized dynamic tags, return the personalized dynamic tags, each initial media data and each static display tag to the terminal corresponding to the display request and display them, so that the searcher can directly select the dynamic tags to search for media data.

[0171] Step 202: Obtain the information recommendation request. The information recommendation request carries the selected media data tags and includes the following steps:

[0172] The system retrieves information recommendation requests sent by the terminal through the media data search page. These requests carry selected media data tags, which are determined from various static display tags.

[0173] Specifically, after the terminal displays various static display tags on the media data search page, the target can select a static display tag through the terminal. When a static display tag is selected, the terminal receives a search instruction for the selected media data tag. The terminal responds to the search instruction by sending an information recommendation request to the server. The server receives the information recommendation request sent by the terminal through the media data search page, parses the information recommendation request, and obtains the selected media data tag.

[0174] Step 210: Recommend the target dynamic tag information to the terminal corresponding to the request, so as to display the target dynamic tag through the terminal, including the following steps:

[0175] The target dynamic tag and each recommended media data are returned to the terminal corresponding to the recommendation request, so that the target dynamic tag can be displayed through the media data search page in the terminal, and each initial media data is updated to each recommended media data and displayed.

[0176] Specifically, the server simultaneously returns the target dynamic tag and each recommended media data to the terminal corresponding to the information recommendation request. Upon receiving the target dynamic tag and the recommended media data, the terminal displays the target dynamic tag on the media data search page. Simultaneously, it updates the initial media data to the recommended media data and displays them. That is, the terminal only displays the recommended media data on the media data search page, not the initial media data. The displayed media data can be either the media data identifier or the media data itself. For example, when the media data is a video, the video name can be displayed; when the media data is an image, the image can be displayed directly.

[0177] In the above embodiments, by obtaining the display request of the media data search page, obtaining each static display tag based on the display request, and obtaining each initial media data based on the target object and each static display tag, the initial media data and each static display tag are returned to the terminal corresponding to the display request, so that the terminal can display each static display tag and each initial media data on the media data search page, which facilitates the target object to search for media data through the media data search page, improves efficiency, and enhances the user experience of the target object.

[0178] In one embodiment, step 204, determining each candidate tag from the tags associated with each recommended media data, includes the following steps:

[0179] Obtain the interest tags corresponding to the target object, and determine each candidate tag based on the interest tags and the tags associated with each recommended media data.

[0180] Interest tags refer to tags that the target audience is interested in. These interest tags can be obtained by creating a profile of the target audience or by determining tags associated with the target audience's historical media viewing data.

[0181] Specifically, the server can retrieve the interest tags corresponding to the target object from the database, or it can retrieve the interest tags corresponding to the target object uploaded by the terminal. Alternatively, the server can retrieve the interest tags corresponding to the target object from the business service provider. Then, the server filters candidate tags based on the interest tags and the tags associated with the recommended media data. Among these, tags that coexist with the tags associated with the recommended media data and the interest tags can be used as candidate tags.

[0182] In one embodiment, the server can directly use the interest tags corresponding to the target object as candidate tags. Furthermore, static display tags among the interest tags can be filtered out, and the remaining interest tags can be used as candidate tags, thereby improving the efficiency of obtaining candidate tags.

[0183] In the above embodiments, each candidate tag is determined by using the interest tags corresponding to the target object, so that the determined tags can satisfy the interests of the target object, thereby improving the accuracy of the determined candidate tags.

[0184] In one embodiment, such as Figure 7 As shown, in step 210, after determining the target dynamic tag from each candidate tag based on the tag recommendation degree, returning the target dynamic tag to the terminal corresponding to the information recommendation request, and displaying the target dynamic tag through the terminal, the method further includes:

[0185] Step 702: Obtain the target information recommendation request, which carries the selected media data tags and the selected dynamic tags.

[0186] Step 704: Based on the target information recommendation request, obtain the target recommendation media data associated with the selected dynamic tags from each recommendation media data according to the preset association relationship.

[0187] Step 706: Determine each target candidate tag based on the selected media data tags and selected dynamic tags, calculate the target tag recommendation degree corresponding to each target candidate tag, and determine each dynamic tag to be displayed from each target candidate tag based on the target tag recommendation degree;

[0188] Step 708: Return the recommended media data for each target and the dynamic tags to be displayed to the terminal corresponding to the information recommendation request, so as to display the recommended media data for each target and the dynamic tags to be displayed through the terminal.

[0189] Among them, "selected dynamic tags" refers to the dynamic tags chosen for media data search. "Target recommended media data" refers to the media data associated with the selected dynamic tags in each recommended media data set. The tags associated with the target recommended media data include both selected media data tags and selected dynamic tags. "Dynamic tags to be displayed" are the dynamic tags recommended to the terminal based on the selected selected media data tags and selected dynamic tags.

[0190] Specifically, after the terminal displays various target dynamic tags, the target object can be the target dynamic tag selected from these tags for media data search. When the terminal receives a search instruction for a target dynamic tag, it sends a target information recommendation request to the server. This request carries the selected media data tag and the selected dynamic tag. When the server receives the target information recommendation request, it responds to the request and retrieves the target recommended media data associated with the selected dynamic tag from the recommended media data according to a preset association relationship. Simultaneously, the server uses the selected media data tag and the selected dynamic tag as the currently selected media data tag, and determines each target candidate tag from the tags associated with each target recommended media data. Based on the preset association relationship, it determines the target candidate media dataset associated with each target candidate tag. The server calculates the tag recommendation degree based on the target recommended media data and the target candidate media dataset associated with each target candidate tag, obtaining the target tag recommendation degree corresponding to each target candidate tag. Based on the target tag recommendation degree, it determines each dynamic tag to be displayed from the target candidate tags. The server will request the corresponding terminal to recommend the media data for each target and the information returned by each dynamic tag to be displayed, so that the media data for each target can be displayed on the terminal.

[0191] In the above embodiments, the target recommended media data associated with the selected dynamic tags are obtained from the recommended media data according to the preset association relationship through the target information recommendation request, and each dynamic tag to be displayed is determined. The target recommended media data and each dynamic tag to be displayed are returned to the terminal corresponding to the information recommendation request so that the target recommended media data and each dynamic tag to be displayed can be displayed through the terminal. Furthermore, the target recommended media data and each dynamic tag to be displayed are obtained by selecting dynamic tags, thereby improving the efficiency and accuracy of obtaining the target recommended media data and each dynamic tag to be displayed.

[0192] In one embodiment, such as Figure 8 As shown, an information display method is provided, which can be applied to... Figure 1 Taking a terminal as an example, it can be understood that this method can also be applied to a server, and to a system that includes both a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0193] Step 802: Display each static display tag on the media data search page.

[0194] The media data search page is the page for searching media data. This page displays tags used for searching media data, including static and dynamic tags. It also displays media data obtained through tag searches. Static tags are tags that are permanently displayed on the media data search page; their content does not change.

[0195] Specifically, the user accesses the media data search page using a terminal. Upon detecting access to the media data search page, the terminal retrieves various static display tags from the server and then displays these static display tags on the media data search page. In one embodiment, the terminal can also retrieve the media data associated with each static display tag from the server simultaneously, and then display the media data associated with each static display tag concurrently.

[0196] Step 804: In response to the selection event of the target static display tag among the various static display tags, display the target dynamic tag in the media data search page; wherein, the target dynamic tag is determined by identifying each candidate tag among the tags corresponding to each recommended media data associated with the target static display tag, calculating the recommendation degree based on each recommended media data associated with the target static display tag and the candidate media datasets associated with each candidate tag respectively, obtaining the tag recommendation degree corresponding to each candidate tag, and determining it from each candidate tag based on the tag recommendation degree, and the tag is used to characterize the attribute information of the media data.

[0197] The target dynamic tags can be obtained through any of the methods described in the above-described information recommendation methods. For example, the target dynamic tags can be determined by identifying candidate tags from the tags corresponding to the recommended media data associated with the target static display tag, calculating the recommendation degree based on the recommended media data associated with the target static display tag and the candidate media datasets associated with each candidate tag, obtaining the tag recommendation degree corresponding to each candidate tag, and determining the tags from the candidate tags based on the tag recommendation degree.

[0198] Specifically, when the terminal detects that a searcher has selected a target static display tag on the media data search page (this selection can be a click, swipe, voice command, press, etc.), the terminal generates an information recommendation request in response to the selection event of the target static display tag among the various static display tags and sends it to the server. Upon receiving the information recommendation request, the server can use any of the above-described information recommendation methods to obtain the target dynamic tag to be recommended and return it to the terminal. The terminal receives the returned target dynamic tag and displays it on the media data search page.

[0199] In one embodiment, the server can filter to obtain a target number of target dynamic tags according to a pre-set quantity. The terminal then obtains the target number of target dynamic tags and displays them on the media data search page. Alternatively, the target number can be determined based on the size of the dynamic tag display area on the media data search page. The number of dynamic tags that can be displayed on the media data search page is calculated based on the size of the target dynamic tag display area and the size of the dynamic tag display area on the media data search page. The terminal then obtains and displays a maximum of the target number of dynamic tags that can be displayed on the media data search page.

[0200] The aforementioned information display method involves displaying various static display tags on the media data search page; and, in response to the selection event of a target static display tag among these tags, displaying a target dynamic tag on the media data search page. The target dynamic tag is determined by identifying candidate tags from the tags corresponding to the recommended media data associated with the target static display tag, calculating the recommendation score for each candidate tag based on the recommended media data associated with the target static display tag and the candidate media datasets associated with each candidate tag, and then determining the tag recommendation score from among the candidate tags. Tags are used to characterize the attribute information of media data. By determining each candidate tag when the target static display tag is selected, calculating the recommendation score of the candidate tags, determining the target dynamic tag based on the recommendation score, and then displaying the target dynamic tag, the accuracy of the displayed tags is improved. This avoids the use of fixed tags, and dynamic tags can be used for media data searching, thereby improving the precision of media data searching.

[0201] In a specific embodiment, such as Figure 9 The diagram illustrates a media data search page, which is used for video searches. This video search page includes a text input search section and a tag search section. Static tags are displayed within the dashed box 910 of the tag search section, and static tag 912 is selected within the dashed box 910. Then, the terminal displays various target dynamic tags within the dashed box 920 of the tag search section. These target dynamic tags are obtained through information recommendation using any of the above-described information recommendation methods based on the static tags 912, thereby improving the accuracy of tag display. Further selection of target dynamic tags for video searching further enhances the precision of the video search.

[0202] In one embodiment, step 804, in response to a selection event for a target static display tag among various static display tags, displays a target dynamic tag in the media data search page, including the steps of:

[0203] In response to the selection event of the target static display tag among the various static display tags, the sub-tags corresponding to the target static display tag are displayed on the media data search page; in response to the selection event of the target sub-tag among the various sub-tags, the various sub-dynamic tags are displayed on the media data search page; wherein, each sub-dynamic tag is determined by identifying each target candidate tag from the tags corresponding to the various recommended media data that are commonly associated with the target static display tag and the target sub-tag, calculating the recommendation degree based on the candidate media datasets associated with the commonly associated recommended media data and each target candidate tag respectively, obtaining the target tag recommendation degree corresponding to each target candidate tag, and determining it from each target candidate tag based on the target tag recommendation degree.

[0204] Sub-tags refer to secondary tags associated with the target static display tag, used to further distinguish the media data associated with the target static display tag. Different static display tags are associated with different sub-tags. Different media data can be associated with the same sub-tag or different sub-tags, and the same media data can be associated with multiple sub-tags. Sub-tags are also static display tags, which are pre-set and fixed tags. The target sub-tag is the selected sub-tag. Sub-dynamic tags refer to the target dynamic tags obtained by recommending information based on the target static display tag and the target sub-tag.

[0205] Specifically, in response to the selection event of the target static display tag among the various static display tags, the sub-tags corresponding to the target static display tag are displayed on the media data search page; in response to the selection event of the target sub-tag among the various sub-tags, the various sub-dynamic tags are displayed on the media data search page; wherein, the various sub-dynamic tags can also be obtained by information recommendation through any of the methods in the above-described information recommendation methods. For example, the various sub-dynamic tags can be determined by identifying each target candidate tag from the tags corresponding to the various recommended media data that are commonly associated with the target static display tag and the target sub-tag, calculating the recommendation degree based on the candidate media datasets that are commonly associated with the various recommended media data and the target candidate tags respectively, obtaining the target tag recommendation degree corresponding to each target candidate tag, and determining it from the various target candidate tags based on the target tag recommendation degree.

[0206] In a specific embodiment, such as Figure 10 The diagram illustrates a media data search page, used for searching videos. This page includes a text input search section and a tag search section. Within the tag search section, dashed box 1010 displays various static display tags. Static display tag 1012 within dashed box 1010 is selected, i.e., "Static Display Tag A". Dashed box 1020 displays the corresponding sub-tags for the static display tags. Users can then select "All" from these sub-tags. The terminal retrieves various target dynamic tags from the server using the target static tags and target sub-tags, and then displays these target dynamic tags (M, N, Q, P, J, L, O) within the dashed box 1030 of the tag search section on the video search page.

[0207] In the above embodiments, by displaying each sub-tag corresponding to the target static tag after selecting it, and then displaying each sub-dynamic tag in the media data search page when the target sub-tag is selected, the accuracy of tag display is further improved. Then, the target dynamic tags can be used for video search, thereby improving the accuracy of video search.

[0208] In one embodiment, after step 804, and after displaying the target dynamic tag in the media data search page in response to a selection event for the target static display tag among the various static display tags, the step further includes:

[0209] In response to the selection event of a target dynamic tag, the target dynamic tag is displayed in the tag adjustment area of ​​the media data search page, each dynamic tag to be displayed is displayed in the dynamic tag display area of ​​the media data search page, and each target recommended media data associated with the target dynamic tag is displayed in the data display area of ​​the media data search page. Each dynamic tag to be displayed is obtained by recommending information using the target static display tag and the target dynamic tag as selected media data tags, and each target recommended media data associated with the target dynamic tag is determined from the recommended media data associated with the target static display tag.

[0210] The label adjustment area is used for dynamic label adjustment, allowing selected dynamic labels to be displayed or deleted. The data display area is used to display media data. The dynamic label display area displays dynamic labels that change depending on the selected label. Selected labels can include both dynamic and static labels. Each dynamic label to be displayed is obtained by recommending information using target static labels and target dynamic labels as selected media data labels. In other words, each dynamic label to be displayed can be obtained by recommending information using any of the above-described information recommendation methods, with target static labels and target dynamic labels as selected media data labels.

[0211] Specifically, in response to a selection event for a target dynamic tag, the terminal generates a media data recommendation request and sends it to the server. Upon receiving the request, the server parses it to obtain the target dynamic tag and then retrieves the target recommended media data corresponding to that tag from the recommended media data. The server then returns each target recommended media data set to the terminal. Upon receiving the target recommended media data from the server, the terminal displays the target dynamic tag in the tag adjustment area of ​​the media data search page, displays each dynamic tag to be displayed in the dynamic tag display area of ​​the media data search page, and displays each target recommended media data set associated with the target dynamic tag in the data display area of ​​the media data search page. The tags associated with this target recommended media data include both target static tags and target dynamic tags.

[0212] In a specific embodiment, such as Figure 11The diagram illustrates a media data search page, used for searching videos. This page includes a text input search section and a tag search section. Within the tag search section, dashed box 1010 displays various static tags, with static tag 1012 selected. Dashed box 1020 displays the sub-tags corresponding to the static tags, with the sub-tag "All" selected. The user can then select "Dynamic Tag M" from the various target dynamic tags. The terminal then displays the selected target dynamic tag "Dynamic Tag M" in the tag adjustment area 1120 as an adjustable tag, and simultaneously displays the corresponding video clips (A, B, C, D, E, F, G, H) in the media data display area 1140. The dynamic tag display area 1130 on the media data search page displays the dynamic tags to be displayed (a, c, O, P, d, e). Specifically, when a selected target sub-tag is displayed in the tag adjustment area, the selected target dynamic tag can be deleted or retained in the dynamic tag display area. This improves the flexibility of tag display and enhances the accuracy of video search.

[0213] In the above embodiments, by selecting the target static display tag and then selecting the target dynamic tag after displaying the target dynamic tag, the target dynamic tag can be displayed in the tag adjustment area of ​​the media data search page in response to the selection event of the target dynamic tag, and the target recommended media data associated with the target dynamic tag can be displayed in the data display area of ​​the media data search page.

[0214] In one embodiment, the target dynamic tag includes at least two; in response to a selection event of the target dynamic tag, the target dynamic tag is displayed in the tag adjustment area of ​​the media data search page, and the various target recommended media data associated with the target dynamic tag are displayed in the data display area of ​​the media data search page, including the steps of:

[0215] In response to the selection event of at least two target dynamic tags, at least two target dynamic tags are displayed in the tag adjustment area of ​​the media data search page, each target dynamic tag to be displayed is displayed in the dynamic tag display area of ​​the media data search page, and each target recommended media data associated with the at least two target dynamic tags is displayed in the data display area of ​​the media data search page. Each target dynamic tag to be displayed is obtained by recommending information using the target static display tag and the at least two target dynamic tags as selected media data tags, and each target recommended media data associated with the at least two target dynamic tags is determined from the recommended media data associated with the target static display tag.

[0216] The dynamic tag display area is used to display dynamic tags. The dynamic tags in this area change depending on the selected tag. The selected tags can include both dynamic tags and static display tags. Each target dynamic tag to be displayed is obtained by recommending information using the target static display tag and at least two target dynamic tags as selected media data tags. That is, each target dynamic tag to be displayed can be obtained by recommending information using any of the above-described information recommendation methods using the target static display tag and at least two target dynamic tags as selected media data tags.

[0217] Specifically, when a searcher selects at least two target dynamic tags consecutively on the terminal, the terminal, in response to the selection event for at least two target dynamic tags, sends a media data search request to the server. This media data search request carries at least two target dynamic tags. Upon receiving the media data search request, the server searches for commonly associated target recommended media data from the recommended media data associated with the target static display tags, based on the at least two target dynamic tags carried. That is, the tags associated with the obtained target recommended media data include all selected tags, including static display tags and all dynamic tags. Then, the server uses the target static display tags and at least two target dynamic tags as selected media data tags and performs information recommendation using any of the above-described information recommendation methods to obtain various target dynamic tags to be displayed. Then, the commonly associated target recommended media data and the target dynamic tags to be displayed are returned to the terminal. The terminal displays at least two target dynamic tags in the tag adjustment area of ​​the media data search page, displays each target dynamic tag to be displayed in the dynamic tag display area of ​​the media data search page, and displays the target recommended media data commonly associated with at least two target dynamic tags in the data display area of ​​the media data search page.

[0218] In a specific embodiment, such as Figure 12The diagram shows a media data search page, which is used to search for videos. The video search page includes a text input search section and a tag search section. Within the tag search section, dashed box 1010 displays various static display tags, with "Static Display Tag A" 1012 being the selected static display tag. Dashed box 1020 displays the sub-tags corresponding to "Static Display Tag A". The "All" sub-tag within dashed box 1020 is the selected target sub-tag. The terminal displays various target dynamic tags within dashed box 1240. The searcher then consecutively selects "Dynamic Tag M" and "Dynamic Tag a". The terminal displays these two target dynamic tags, "Dynamic Tag M" and "Dynamic Tag a", in the tag adjustment area 1220, making them adjustable tags. The dynamic tag display area 1240 on the media data search page displays various target dynamic tags (F, W, X, U, d, N) to be displayed, while the corresponding video clips (a, b, c, d, e, f, g, h) are displayed in the media data display area 1260. When a selected target sub-tag is displayed in the tag adjustment area, the selected target dynamic tag can be deleted from the dynamic tag display area. This improves the flexibility of tag display and enhances the accuracy of video search.

[0219] In one embodiment, after selecting a target dynamic tag and completing the media data display, the searcher can continue selecting target dynamic tags. At this time, the tag adjustment area on the terminal displays two target dynamic tags, the dynamic tag display area shows the dynamic tags obtained through information recommendation after selecting the tag, and the data display area displays the target recommended media data commonly associated with the two target dynamic tags. This target recommended media data is determined from the recommended media data associated with the target static display tag.

[0220] In the above embodiments, in response to the selection event of at least two target dynamic tags, at least two target dynamic tags are displayed in the tag adjustment area of ​​the media data search page, each target dynamic tag to be displayed is displayed in the dynamic tag display area of ​​the media data search page, and each target recommended media data associated with at least two target dynamic tags is displayed in the data display area of ​​the media data search page, which can further improve the accuracy of the obtained target recommended media data.

[0221] In one embodiment, after displaying target dynamic tags in the tag adjustment area of ​​the media data search page and displaying various target recommended media data associated with the target dynamic tags in the data display area of ​​the media data search page, the method further includes the following steps:

[0222] In response to a deletion event of a target dynamic tag in the tag adjustment area, the target dynamic tag is deleted from the tag adjustment area, and the recommended media data associated with the target static display tag is displayed in the data display area.

[0223] The delete event is used to delete the target dynamic label in the label adjustment area. It can be implemented by responding to the delete operation, which includes but is not limited to click operation, press operation, voice operation, drag operation, swipe operation, etc.

[0224] Specifically, when a searcher needs to delete a target dynamic tag in the tag adjustment area, they can do so through a deletion operation. In this case, the terminal responds to the deletion event for the target dynamic tag in the tag adjustment area, deletes the target dynamic tag in the tag adjustment area, and displays the recommended media data associated with the target static display tag in the data display area. In one embodiment, when there are two or more target dynamic tags in the tag adjustment area, the terminal responds to the deletion event for the target dynamic tag in the tag adjustment area and deletes the target dynamic tag in the tag adjustment area. At this time, the tag adjustment area displays the remaining target dynamic tags, and the data display area displays the recommended media data associated with the remaining target dynamic tags. Simultaneously, the dynamic tag display area displays dynamic tags obtained by recommending information using the selected static display tag and the remaining target dynamic tags as selected media data tags. The recommended media data associated with the remaining target dynamic tags are also determined from the recommended media data associated with the target static display tags.

[0225] In a specific embodiment, it can be Figure 12 Delete any tag within the tag adjustment area 1220. For example, when deleting "Dynamic Tag a", the media search page displayed on the terminal could be as follows: Figure 11 The page shown displays only "Dynamic Tag M" in the tag adjustment area, various dynamic tags obtained by recommending information using "Static Tag A", "All" and "Dynamic Tag M" as selected media data tags in the dynamic tag display area, and video clips (A, B, C, D, E, F, G, H) associated with the selected media data tags in the data display area.

[0226] In the above embodiments, by responding to a deletion event of a target dynamic tag in the tag adjustment area, the target dynamic tag is deleted in the tag adjustment area, and the recommended media data associated with the target static display tag is displayed in the data display area. This achieves the adjustment of the target dynamic tag in the tag adjustment area, further improving the accuracy of tag display and media data search.

[0227] In one embodiment, the data display method further includes the step of:

[0228] In response to the selection event of historically selected static display tags in each static display tag, the historical dynamic tags are displayed in the media data search page, and the historical target dynamic tags are displayed in the tag adjustment area. The historical target dynamic tags and each historical dynamic tag are recorded when the historically selected static display tags are deselected.

[0229] Among them, "historically selected static display tags" refers to static display tags that were selected before the currently selected target static display tag. "Historically selected dynamic tags" refers to dynamic tags obtained after recommending information by using historically selected static display tags as target static display tags. "Historically selected target dynamic tags" refers to historical dynamic tags selected from various historical dynamic tags. Historically selected target dynamic tags and each historical dynamic tag are recorded when a historically selected static display tag is deselected. The recording time is the time when a historically selected static display tag is deselected. This deselection time can be any time after selecting a historically selected static display tag; for example, the deselection time could be the time when another static display tag is selected after selecting a historically selected static display tag, or the time when the user exits the media data search page after selecting a historically selected static display tag.

[0230] Specifically, when the terminal detects that a target static display tag has been selected, it records the information displayed on the media data search page before the target static display tag was selected, including previously selected static display tags, various historical dynamic tags, and historical target dynamic tags displayed in the tag adjustment area. It can also record the historical target recommended media data corresponding to the historical target dynamic tags. After currently selecting a target static display tag, the terminal can reselect a static display tag from the various static display tags. When the reselected static display tag is a previously selected static display tag, the terminal responds to the selection event of the previously selected static display tag among the various static display tags, obtains the recorded various historical dynamic tags and historical target dynamic tags, and then displays the various historical dynamic tags on the media data search page and the historical target dynamic tags in the tag adjustment area. In one embodiment, it can also obtain the various historical target recommended media data associated with the historical target dynamic tags and display the various historical target recommended media data in the data display area of ​​the media data search page.

[0231] In the above embodiments, by responding to the selection event of the historically selected static display tags in each static display tag, the recorded historical dynamic tags and historical target dynamic tags are obtained, and then each historical dynamic tag is displayed on the media data search page, and the historical target dynamic tag is displayed in the tag adjustment area. This can restore the state of historical tag selection without requiring the searcher to reselect, thereby improving the efficiency of tag selection and further improving the efficiency of data search.

[0232] In a specific embodiment, such as Figure 13 As shown, an information recommendation method is provided, taking server-side execution as an example, and specifically includes the following steps:

[0233] Step 1302: Obtain the display request of the media data search page, with the display request carrying the target object; obtain each static display tag based on the display request, and obtain each initial media data based on the target object and each static display tag.

[0234] Step 1304: Return each initial media data and each static display tag to the terminal corresponding to the display request, so that the terminal can display each static display tag and each initial media data on the media data search page; obtain the information recommendation request sent by the terminal through the media data search page, the information recommendation request carries the selected media data tag, and the selected media data tag is determined from each static display tag.

[0235] Step 1306: Obtain the recommended media data associated with the selected media data tags according to the preset association relationship between media data and tags, and obtain the tags associated with each recommended media data according to the preset association relationship. Filter the selected media data tags from the tags associated with each recommended media data to obtain each initial candidate tag.

[0236] Step 1308: When there is no preset static display tag among the initial candidate tags, each initial candidate tag is used as a candidate tag; when there is a preset static display tag among the initial candidate tags, the preset static display tags among the initial candidate tags are filtered to obtain each candidate tag.

[0237] Step 1310: Based on preset association relationships, determine the candidate media datasets associated with each candidate tag from each recommended media data. Calculate the number of recommendations corresponding to each recommended media data, and calculate the number of candidates in each candidate media dataset associated with each candidate tag. Calculate the ratio of the number of candidates to the number of recommendations for each candidate tag to obtain the tag recommendation degree for each candidate tag.

[0238] Step 1312: Based on the recommendation degree of the tags, select from each candidate tag sequentially to obtain at least two target dynamic tags. Obtain each similar tag pair, and based on each similar tag pair, filter out dissimilar tags from the at least two dynamic tags to obtain each dissimilar dynamic tag.

[0239] Step 1314: Return each dissimilar dynamic tag and each recommended media data to the terminal corresponding to the information recommendation request, so as to display each dissimilar dynamic tag through the media data search page in the terminal, and update each initial media data to each recommended media data and display it.

[0240] Step 1316: Obtain the target information recommendation request. The target information recommendation request carries the selected media data tags and the selected dynamic tags. Based on the target information recommendation request, retrieve the target recommended media data associated with the selected dynamic tags from each recommended media data according to the preset association relationship, and return each target recommended media data to the terminal corresponding to the information recommendation request so that the target recommended media data can be displayed on the terminal.

[0241] In a specific embodiment, such as Figure 14 The diagram illustrates the overall architecture of an information recommendation method applied to an online video media platform. Specifically:

[0242] Users can access online video media platforms to search for videos. When searching, users can access the search page through the platform, for example, by clicking the search box. The search page displays manually configured static tags, along with popular movies and TV shows under each static tag. It can also display personalized dynamic tags tailored to the searcher's interest. Users can select and delete tags. For example, when a user selects a static tag, the tag recommendation server receives a recommendation request, retrieves candidate tags based on the selected static tag, calculates the recommendation score for each candidate tag, sorts them according to their recommendation score, and selects the top-ranked candidate tags to obtain dynamic tags. These dynamic tags are then filtered to identify dissimilar dynamic tags, which are then returned to the terminal for display.

[0243] The following procedure can be used to filter dissimilar tags:

[0244]

[0245]

[0246] The input consists of a sorted tag set S[t1...tn], a selected tag set A, and a set of highly similar tag pairs B. The output is a set of dissimilar target dynamic tags V. The process involves adding the dynamic tags sorted by recommendation level to the final dynamic tag recommendation set sequentially. Each time a tag is added, it checks if a similar dynamic tag has already been added. If so, the tag is not added; otherwise, it continues to be added until all dynamic tags have been traversed. The final tag recommendation set consists of all dissimilar dynamic tags.

[0247] Then, the searcher can select a dynamic tag to display on the search page. The terminal responds to the selection event for the target dynamic tag by displaying the target dynamic tag in the tag adjustment area of ​​the media data search page, and simultaneously sends an information recommendation request to the server. The server retrieves various target recommended movies and TV shows based on the selected static display tag and the target dynamic tag, and returns these recommended movies and TV shows to the terminal for display. The searcher can also delete the selected target dynamic tag in the tag adjustment area of ​​the search page, restoring the search page to the state before the deleted target dynamic tag was selected. The server can also sort the movies and TV shows to be sent to the terminal using a movie and TV show recall and sorting service. When the terminal receives the returned movies and TV shows, they are displayed in order. The terminal can obtain the names and corresponding theme images of each movie and TV show associated with the selected target dynamic tag from the server. The terminal displays the names and corresponding images of each movie and TV show, and the searcher can scroll up and down on the search page to view the movies and TV shows under the currently selected tag.

[0248] The search page can also switch to a new static display tab. In this case, the terminal first records the current selection state of the static display tab, i.e., the information currently displayed on the search page, including the selected static display tab, the various dynamic tabs displayed, and the dynamic tabs displayed in the tab adjustment area. Then, the terminal selects a new static display tab. If this new static display tab is a historical target static display tab, the recorded selection state of the new static display tab is directly retrieved, and the search page is restored to the state it was in when the new static display tab was selected—that is, the previously saved selection state is restored. The searcher can then browse the movies and TV shows under the currently selected tab on the restored search page.

[0249] In one embodiment, a comparative test can be conducted between using the information display method and the information recommendation method of this application and not using the information display method and the information recommendation method of this application, as shown in Table 1 below, which is a schematic diagram of the comparative test results.

[0250]

[0251] Among them, the indicators obtained by using the information display method and information recommendation method of this application during the comparative test showed a significant increase compared with the indicators obtained by not using the information display method and information recommendation method of this application.

[0252] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0253] Based on the same inventive concept, this application also provides an information recommendation device and an information display device for implementing the information recommendation method and information display method described above. The solution provided by this device is similar to the implementation scheme described in the above methods. Therefore, the specific limitations of one or more information display devices and information recommendation devices provided below can be found in the limitations of the information display method and information recommendation method described above, and will not be repeated here.

[0254] In one embodiment, such as Figure 15 As shown, an information recommendation device 1500 is provided, including: a request acquisition module 1502, a candidate tag determination module 1504, a data determination module 1506, a recommendation degree calculation module 1508, and a tag recommendation module 1510, wherein:

[0255] The request acquisition module 1502 is used to acquire information recommendation requests, which carry the selected media data tags.

[0256] The candidate tag determination module 1504 is used to obtain each recommended media data associated with the selected media data tags according to the preset association relationship between media data and tags, and to determine each candidate tag from the tags associated with each recommended media data. The tags are used to characterize the attribute information of the media data.

[0257] The data determination module 1506 is used to determine the candidate media dataset associated with each candidate tag from each recommended media data based on a preset association relationship;

[0258] The recommendation degree calculation module 1508 is used to calculate the tag recommendation degree based on each recommendation media data and the candidate media dataset associated with each candidate tag, and obtain the tag recommendation degree corresponding to each candidate tag.

[0259] The tag recommendation module 1510 is used to determine the target dynamic tag from each candidate tag based on the tag recommendation degree, and return the target dynamic tag to the terminal corresponding to the information recommendation request so that the target dynamic tag can be displayed on the terminal.

[0260] In one embodiment, the candidate tag determination module 1504 is further configured to obtain the tags associated with each recommended media data according to a preset association relationship, filter the selected media data tags from the tags associated with each recommended media data to obtain each initial candidate tag; when there is no preset static display tag among the initial candidate tags, the initial candidate tags are used as each candidate tag; when there is a preset static display tag among the initial candidate tags, the preset static display tags present in the initial candidate tags are filtered to obtain each candidate tag.

[0261] In one embodiment, the information recommendation device 1500 further includes:

[0262] The reminder module is used to generate a reminder message indicating that a dynamic tag does not exist when all initial candidate tags are preset static display tags. The reminder message is then returned to the terminal for display.

[0263] In one embodiment, the data determination module 1506 is further configured to determine the current candidate label from each candidate label, filter the recommended media data associated with the current candidate label from each recommended media data according to a preset association relationship, and obtain the candidate media dataset associated with the current candidate label; and traverse each candidate label to obtain the candidate media dataset associated with each candidate label respectively.

[0264] In one embodiment, the recommendation calculation module 1508 is further configured to calculate the number of recommendations corresponding to each recommended media data, and calculate the number of candidates in the candidate media dataset associated with each candidate tag; calculate the ratio of the number of candidates corresponding to each candidate tag to the number of recommendations, and obtain the tag recommendation degree corresponding to each candidate tag.

[0265] In one embodiment, the recommendation calculation module 1508 is further configured to obtain the data quality information corresponding to each recommended media data, and obtain the candidate quality information corresponding to the candidate media data in the candidate media dataset associated with each candidate tag; perform a sum calculation based on the data quality information corresponding to each recommended media data to obtain the total data quality information, and perform a sum calculation based on the candidate quality information corresponding to the candidate media data in the candidate media dataset to obtain the total candidate quality information corresponding to each candidate tag; and obtain the tag recommendation degree corresponding to each candidate tag based on the total candidate quality information and the total data quality information.

[0266] In one embodiment, the tag recommendation module 1510 includes:

[0267] The selection unit is used to select from each candidate tag in turn based on the tag recommendation degree to obtain at least two target dynamic tags;

[0268] The filtering unit is used to obtain each pair of similar tags, and to filter out dissimilar tags from at least two dynamic tags based on each pair of similar tags to obtain each dissimilar tag.

[0269] The return unit is used to recommend the terminal corresponding to the return information of each dissimilar tag, so that the terminal can display each dissimilar tag.

[0270] In one embodiment, the filtering unit is further configured to obtain each set label, pair each set label to obtain each label pair, obtain the co-occurrence count of each label pair, and regard label pairs whose co-occurrence count exceeds a preset threshold as similar label pairs.

[0271] In one embodiment, the filtering unit is further configured to acquire each set label, pair each set label to obtain each label pair, calculate the point-to-point mutual information corresponding to each label pair, and classify the label pairs whose point-to-point mutual information exceeds a preset information threshold as similar label pairs.

[0272] In one embodiment, the filtering unit is further configured to obtain each set label and the co-occurrence information corresponding to each set label, establish a label relationship graph based on each set label and the co-occurrence information corresponding to each set label, select label node pairs from the label relationship graph, calculate the density of the label node pairs based on the label relationship graph, and regard the label node pairs with a density exceeding a preset density threshold as similar label pairs. The density is calculated based on the connection edge weight between the two label nodes in the label node pair and the outgoing edge weight of each label node.

[0273] In one embodiment, the information recommendation device 1500 further includes:

[0274] The page display module is used to obtain the display request of the media data search page, and the display request carries the target object; based on the display request, it obtains each static display tag, and based on the target object and each static display tag, it obtains each initial media data, and returns each initial media data and each static display tag to the terminal corresponding to the display request, so that the terminal can display each static display tag and each initial media data on the media data search page;

[0275] The request acquisition module 1502 is also used to acquire information recommendation requests sent by the terminal through the media data search page. The information recommendation requests carry selected media data tags, which are determined from various static display tags.

[0276] The tag recommendation module 1510 is also used to return the target dynamic tag and each recommended media data to the terminal corresponding to the information recommendation request, so as to display the target dynamic tag through the media data search page in the terminal, and update each initial media data to each recommended media data and display it.

[0277] In one embodiment, the candidate tag determination module 1504 is further configured to obtain the interest tags corresponding to the target object, and determine each candidate tag based on the interest tags and the tags associated with each recommended media data.

[0278] In one embodiment, the information recommendation device 1500 further includes:

[0279] The dynamic tag selection module is used to obtain target information recommendation requests, which carry selected media data tags and selected dynamic tags. Based on the target information recommendation requests, the module retrieves the target recommended media data associated with the selected dynamic tags from each recommended media data according to preset association relationships. Based on the selected media data tags and selected dynamic tags, the module determines each target candidate tag, calculates the target tag recommendation degree corresponding to each target candidate tag, and determines each dynamic tag to be displayed from each target candidate tag based on the target tag recommendation degree. The module returns each target recommended media data and each dynamic tag to be displayed to the terminal corresponding to the information recommendation request, so that the target recommended media data and each dynamic tag to be displayed can be displayed through the terminal.

[0280] In one embodiment, such as Figure 16 As shown, an information display device 1600 is provided, including: a static label display module 1602 and a dynamic label display module 1604, wherein:

[0281] The static tag display module 1602 is used to display various static tags on the media data search page;

[0282] The dynamic tag display module 1604 is used to display the target dynamic tag on the media data search page in response to the selection event of the target static display tag among the various static display tags. The target dynamic tag is determined by identifying each candidate tag from the tags corresponding to each recommended media data associated with the target static display tag, calculating the recommendation degree based on each recommended media data associated with the target static display tag and the candidate media datasets associated with each candidate tag, and determining the tag recommendation degree from each candidate tag based on the tag recommendation degree. The tag is used to represent the attribute information of the media data.

[0283] In one embodiment, the dynamic tag display module 1604 is further configured to, in response to a selection event of a target static display tag among the various static display tags, display each sub-tag corresponding to the target static display tag on the media data search page; and, in response to a selection event of a target sub-tag among the various sub-tags, display each sub-dynamic tag on the media data search page; each sub-dynamic tag is determined by identifying each target candidate tag from the tags corresponding to each recommended media data that is commonly associated with the target static display tag and the target sub-tag, calculating the recommendation degree based on the candidate media datasets associated with each commonly associated recommended media data and each target candidate tag respectively, obtaining the target tag recommendation degree corresponding to each target candidate tag, and determining it from each target candidate tag based on the target tag recommendation degree.

[0284] In one embodiment, the information display device 1600 further includes:

[0285] The media data display module is used to respond to the selection event of a target dynamic tag. It displays the target dynamic tag in the tag adjustment area of ​​the media data search page, displays each dynamic tag to be displayed in the dynamic tag display area of ​​the media data search page, and displays each target recommended media data associated with the target dynamic tag in the data display area of ​​the media data search page. Each dynamic tag to be displayed is obtained by recommending information using the target static display tag and the target dynamic tag as selected media data tags. The target recommended media data associated with the target dynamic tag is determined from the recommended media data associated with the target static display tag.

[0286] In one embodiment, the target dynamic label includes at least two;

[0287] The media data display module is also used to respond to the selection event of at least two target dynamic tags, display at least two target dynamic tags in the tag adjustment area of ​​the media data search page, display each target dynamic tag to be displayed in the dynamic tag display area of ​​the media data search page, and display each target recommended media data that is commonly associated with the at least two target dynamic tags in the data display area of ​​the media data search page. Each target dynamic tag to be displayed is obtained by recommending information using the target static display tag and at least two target dynamic tags as selected media data tags. The target recommended media data that is commonly associated with the at least two target dynamic tags is determined from the recommended media data associated with the target static display tag.

[0288] In one embodiment, the information display device 1600 further includes:

[0289] The deletion module is used to respond to the deletion event of the target dynamic tag in the tag adjustment area, delete the target dynamic tag in the tag adjustment area, and display the various recommended media data associated with the target static display tag in the data display area.

[0290] In one embodiment, the information display device 1600 further includes:

[0291] The history display module is used to respond to the selection event of the historically selected static display tabs in each static display tab, display each historical dynamic tab in the media data search page, and display the historical target dynamic tab in the tab adjustment area. The historical target dynamic tab and each historical dynamic tab are recorded when the historically selected static display tab is deselected.

[0292] Each module in the aforementioned information recommendation device and information display device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0293] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 17As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores tag data, media data, etc. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network. When executed by the processor, the computer program implements an information recommendation method and an information display method.

[0294] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 18 As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements an information recommendation method and an information display method. The display unit of the computer device is used to form a visually visible image. It can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0295] Those skilled in the art will understand that Figure 17 and Figure 18The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0296] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0297] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0298] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0299] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0300] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0301] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0302] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An information recommendation method, characterized in that, The method includes: Obtain an information recommendation request, wherein the information recommendation request carries selected media data tags; According to the preset association relationship between media data and tags, obtain each recommended media data associated with the selected media data tags, and determine each candidate tag from the tags associated with each recommended media data. The tags are used to characterize the attribute information of the media data. Based on the preset association relationship, determine the candidate media datasets associated with each candidate tag from each of the recommended media data; The tag recommendation score is calculated based on the recommended media data and the candidate media datasets associated with each candidate tag, to obtain the tag recommendation score corresponding to each candidate tag. This includes: obtaining the data quality information corresponding to each recommended media data, and obtaining the candidate quality information corresponding to the candidate media data in the candidate media datasets associated with each candidate tag; performing a summation calculation based on the data quality information corresponding to each recommended media data to obtain total data quality information, and performing a summation calculation based on the candidate quality information corresponding to the candidate media data in the candidate media datasets to obtain total candidate quality information corresponding to each candidate tag; and obtaining the tag recommendation score corresponding to each candidate tag based on the total candidate quality information corresponding to each candidate tag and the total data quality information. Based on the tag recommendation score, a target dynamic tag is determined from each candidate tag, and the target dynamic tag is returned to the terminal corresponding to the information recommendation request so that the target dynamic tag can be displayed through the terminal.

2. The method according to claim 1, characterized in that, The step of determining each candidate tag from the tags associated with each of the recommended media data includes: According to the preset association relationship, obtain the tags associated with each recommended media data, filter the selected media data tags from the tags associated with each recommended media data, and obtain each initial candidate tag; When there is no preset static display tag among the initial candidate tags, the initial candidate tags are used as the candidate tags. When a preset static display tag exists among the initial candidate tags, the preset static display tag is filtered out to obtain the candidate tags.

3. The method according to claim 2, characterized in that, The method further includes: When all the initial candidate tags are preset static display tags, a dynamic tag not found reminder message is generated and returned to the terminal so that the dynamic tag not found reminder message can be displayed on the terminal.

4. The method according to claim 1, characterized in that, The step of determining the candidate media dataset associated with each candidate tag from the various recommended media data based on the preset association relationship includes: The current candidate tag is determined from the candidate tags, and the recommended media data associated with the current candidate tag is filtered from the recommended media data according to the preset association relationship to obtain the candidate media dataset associated with the current candidate tag; By traversing each candidate label, a candidate media dataset associated with each candidate label is obtained.

5. The method according to claim 1, characterized in that, The step of calculating the tag recommendation score based on the candidate media datasets associated with each recommended media data and each candidate tag, to obtain the tag recommendation score corresponding to each candidate tag, includes: Calculate the number of recommendations corresponding to each recommended media data, and calculate the number of candidates in the candidate media dataset associated with each candidate tag; Calculate the ratio of the number of candidates corresponding to each candidate tag to the number of recommendations to obtain the tag recommendation degree corresponding to each candidate tag.

6. The method according to claim 1, characterized in that, The step of determining a target dynamic tag from the candidate tags based on the tag recommendation score, and returning the target dynamic tag to the terminal corresponding to the information recommendation request, so as to display the target dynamic tag through the terminal, includes: Based on the recommendation degree of the tags, at least two target dynamic tags are selected from each of the candidate tags in turn. Obtain each similar tag pair, and based on each similar tag pair, filter out dissimilar tags from the at least two target dynamic tags to obtain each dissimilar tag; The dissimilar tags are returned to the terminal corresponding to the information recommendation request, so that the dissimilar tags can be displayed on the terminal.

7. The method according to claim 6, characterized in that, The process of obtaining each similar tag pair includes: Obtain each set tag, and pair the set tags to obtain tag pairs; Obtain the co-occurrence count of each tag pair, and identify tag pairs whose co-occurrence count exceeds a preset threshold as similar tag pairs.

8. The method according to claim 6, characterized in that, The process of obtaining each similar tag pair includes: Obtain each set tag, and pair the set tags to obtain tag pairs; Calculate the point-to-point mutual information corresponding to each tag pair, and identify tag pairs whose point-to-point mutual information exceeds a preset information threshold as similar tag pairs.

9. The method according to claim 6, characterized in that, The process of obtaining each similar tag pair includes: Obtain each set label and the co-occurrence information corresponding to each set label, and establish a label relationship graph based on each set label and the co-occurrence information corresponding to each set label; Tag node pairs are selected from the tag relationship graph, and the density of the tag node pairs is calculated based on the tag relationship graph. Tag node pairs with a density exceeding a preset density threshold are regarded as similar tag pairs. The density is calculated based on the connection edge weight between the two tag nodes in the tag node pair and the outgoing edge weight of each tag node.

10. The method according to claim 1, characterized in that, Prior to the information recommendation request, the following is also included: Obtain a display request for the media data search page, wherein the display request carries the target object; Based on the display request, obtain each static display tag, and based on the target object and each static display tag, obtain each initial media data; The initial media data and the static display tags are returned to the terminal corresponding to the display request, so that the static display tags and the initial media data can be displayed on the media data search page through the terminal; The information recommendation request, which carries selected media data tags, includes: The system obtains an information recommendation request sent by the terminal through the media data search page, wherein the information recommendation request carries selected media data tags, and the selected media data tags are determined from the various static display tags; The step of returning the target dynamic tag to the terminal corresponding to the information recommendation request, so as to display the target dynamic tag through the terminal, includes: The target dynamic tag and the recommended media data are returned to the terminal corresponding to the information recommendation request, so that the target dynamic tag can be displayed through the media data search page in the terminal, and the initial media data can be updated to the recommended media data and displayed.

11. The method according to claim 10, characterized in that, The step of determining each candidate tag from the tags associated with each of the recommended media data includes: Obtain the interest tags corresponding to the target object, and determine each candidate tag based on the interest tags and the tags associated with each recommended media data.

12. The method according to claim 1, characterized in that, After determining the target dynamic tag from the candidate tags based on the tag recommendation score, and returning the target dynamic tag to the terminal corresponding to the information recommendation request for displaying the target dynamic tag through the terminal, the process further includes: A target information recommendation request is obtained, the target information recommendation request carrying the selected media data tag and the selected dynamic tag, the selected dynamic tag being determined from the target dynamic tag; Based on the target information, the recommendation request retrieves the target recommended media data associated with the selected dynamic tag from the recommended media data according to the preset association relationship; Based on the selected media data tags and the selected dynamic tags, each target candidate tag is determined, the target tag recommendation degree corresponding to each target candidate tag is calculated, and each dynamic tag to be displayed is determined from each target candidate tag based on the target tag recommendation degree; The target recommended media data and the dynamic tags to be displayed are returned to the terminal corresponding to the information recommendation request, so that the target recommended media data and the dynamic tags to be displayed can be displayed through the terminal.

13. An information display method, characterized in that, The method includes: Display various static display tags on the media data search page; In response to a selection event of a target static display tag among the various static display tags, the target dynamic tag is displayed on the media data search page; The target dynamic tag is determined by identifying candidate tags from the tags corresponding to the recommended media data associated with the target static display tag. Recommendation scores are calculated based on the recommended media data associated with the target static display tag and the candidate media datasets associated with each candidate tag, resulting in a tag recommendation score for each candidate tag. The tag is determined from the candidate tags based on the tag recommendation score. The tag is used to characterize the attribute information of the media data. The tag recommendation score for each candidate tag is obtained based on the total candidate quality information and the total data quality information corresponding to each candidate tag. The total candidate quality information is calculated by summing the candidate quality information corresponding to the candidate media data in the candidate media dataset. The total data quality information is calculated by summing the data quality information corresponding to each recommended media data. The candidate media dataset is associated with each candidate tag.

14. The method according to claim 13, characterized in that, The step of displaying a target dynamic tag in the media data search page in response to a selection event of a target static display tag among the various static display tags includes: In response to a selection event of a target static display tag among the various static display tags, the sub-tags corresponding to the target static display tag are displayed on the media data search page; In response to a selection event of a target sub-tag in each of the sub-tags, each sub-dynamic tag is displayed on the media data search page; Each sub-dynamic tag is determined by identifying each target candidate tag from the tags corresponding to each recommended media data that are commonly associated with the target static display tag and the target sub-tag. The recommendation degree is calculated based on the candidate media datasets that are commonly associated with each recommended media data and the target candidate tags respectively, to obtain the target tag recommendation degree corresponding to each target candidate tag, and the target tag recommendation degree is determined from each target candidate tag.

15. The method according to claim 13, characterized in that, After displaying the target dynamic tag in the media data search page in response to a selection event for the target static display tag among the various static display tags, the method further includes: In response to a selection event for the target dynamic tag, the target dynamic tag is displayed in the tag adjustment area of ​​the media data search page, each dynamic tag to be displayed is displayed in the dynamic tag display area of ​​the media data search page, and each target recommended media data associated with the target dynamic tag is displayed in the data display area of ​​the media data search page. Each dynamic tag to be displayed is obtained by recommending information using the target static display tag and the target dynamic tag as selected media data tags, and each target recommended media data associated with the target dynamic tag is determined from the recommended media data associated with the target static display tag.

16. The method according to claim 15, characterized in that, The target dynamic tags include at least two; In response to a selection event for the target dynamic tag, the target dynamic tag is displayed in the tag adjustment area of ​​the media data search page, each dynamic tag to be displayed is displayed in the dynamic tag display area of ​​the media data search page, and each target recommended media data associated with the target dynamic tag is displayed in the data display area of ​​the media data search page, including: In response to the selection event of the at least two target dynamic tags, the at least two target dynamic tags are displayed in the tag adjustment area of ​​the media data search page, each target dynamic tag to be displayed is displayed in the dynamic tag display area of ​​the media data search page, and each target recommended media data commonly associated with the at least two target dynamic tags is displayed in the data display area of ​​the media data search page. The each target dynamic tag to be displayed is obtained by recommending information using the target static display tag and the at least two target dynamic tags as selected media data tags, and the each target recommended media data commonly associated with the at least two target dynamic tags is determined from the recommended media data associated with the target static display tag.

17. The method according to claim 15, characterized in that, The method, in response to a selection event for the target dynamic tag, further includes displaying the target dynamic tag in the tag adjustment area of ​​the media data search page, displaying each dynamic tag to be displayed in the dynamic tag display area of ​​the media data search page, and displaying each target recommended media data associated with the target dynamic tag in the data display area of ​​the media data search page. In response to a deletion event of the target dynamic tag in the tag adjustment area, the target dynamic tag is deleted in the tag adjustment area, and the recommended media data associated with the target static display tag is displayed in the data display area.

18. The method according to claim 15, characterized in that, The method further includes: In response to a selection event of a historically selected static display tag among the various static display tags, the various historical dynamic tags are displayed on the media data search page, and the historical target dynamic tag is displayed in the tag adjustment area. The historical target dynamic tag and the various historical dynamic tags are recorded when the historically selected static display tag is deselected.

19. An information recommendation device, characterized in that, The device includes: The request acquisition module is used to acquire information recommendation requests, wherein the information recommendation requests carry selected media data tags; The candidate tag determination module is used to obtain each recommended media data associated with the selected media data tags according to the preset association relationship between media data and tags, and to determine each candidate tag from the tags associated with each recommended media data, wherein the tag is used to characterize the attribute information of the media data; The data determination module is used to determine the candidate media dataset associated with each candidate tag from each of the recommended media data based on the preset association relationship; The recommendation score calculation module is used to calculate the tag recommendation score based on the recommended media data and the candidate media datasets associated with each candidate tag, to obtain the tag recommendation score corresponding to each candidate tag. This includes: obtaining data quality information corresponding to each recommended media data, and obtaining candidate quality information corresponding to candidate media data in the candidate media datasets associated with each candidate tag; performing a summation calculation based on the data quality information corresponding to each recommended media data to obtain total data quality information, and performing a summation calculation based on the candidate quality information corresponding to candidate media data in the candidate media datasets to obtain total candidate quality information corresponding to each candidate tag; and obtaining the tag recommendation score corresponding to each candidate tag based on the total candidate quality information and the total data quality information. The tag recommendation module is used to determine a target dynamic tag from the candidate tags based on the tag recommendation degree, and return the target dynamic tag to the terminal corresponding to the information recommendation request so as to display the target dynamic tag through the terminal.

20. The apparatus according to claim 19, characterized in that, The candidate tag determination module also obtains the tags associated with each recommended media data according to the preset association relationship, filters the selected media data tags from the tags associated with each recommended media data, and obtains each initial candidate tag. When there is no preset static display tag among the initial candidate tags, the initial candidate tags are used as the candidate tags. When a preset static display tag exists among the initial candidate tags, the preset static display tag is filtered out to obtain the candidate tags.

21. The apparatus according to claim 20, characterized in that, The device further includes: The reminder module is used to generate a reminder message that a dynamic tag does not exist when all the initial candidate tags are preset static display tags, and return the reminder message to the terminal so that the reminder message can be displayed on the terminal.

22. The apparatus according to claim 19, characterized in that, The data determination module is further configured to determine the current candidate tag from each candidate tag, and filter the recommended media data associated with the current candidate tag from each recommended media data according to the preset association relationship, so as to obtain the candidate media dataset associated with the current candidate tag; By traversing each candidate label, a candidate media dataset associated with each candidate label is obtained.

23. The apparatus according to claim 19, characterized in that, The recommendation calculation module is also used to calculate the number of recommendations corresponding to each recommended media data, and to calculate the number of candidates in the candidate media dataset associated with each candidate tag. Calculate the ratio of the number of candidates corresponding to each candidate tag to the number of recommendations to obtain the tag recommendation degree corresponding to each candidate tag.

24. The apparatus according to claim 19, characterized in that, The tag recommendation module includes: The selection unit is used to select from each candidate tag sequentially based on the tag recommendation degree to obtain at least two target dynamic tags; A filtering unit is used to obtain each pair of similar tags, and to filter dissimilar tags from the at least two target dynamic tags based on each pair of similar tags to obtain each dissimilar tag; The return unit is used to return each dissimilar tag to the terminal corresponding to the information recommendation request, so that the dissimilar tags can be displayed through the terminal.

25. The apparatus according to claim 24, characterized in that, The filtering unit is also used to obtain each set label, pair the set labels to obtain each label pair, obtain the co-occurrence count of each label pair, and take the label pairs whose co-occurrence count exceeds a preset threshold as similar label pairs.

26. The apparatus according to claim 24, characterized in that, The filtering unit is also used to acquire each set label, pair the set labels to obtain each label pair, calculate the point-to-point mutual information corresponding to each label pair, and take the label pairs whose point-to-point mutual information exceeds a preset information threshold as similar label pairs.

27. The apparatus according to claim 24, characterized in that, The filtering unit is further configured to acquire each set label and acquire co-occurrence information corresponding to each set label, establish a label relationship graph based on each set label and the co-occurrence information corresponding to each set label; select label node pairs from the label relationship graph, calculate the density of the label node pairs based on the label relationship graph, and take label node pairs with a density exceeding a preset density threshold as similar label pairs. The density is calculated based on the connection edge weight between the two label nodes in the label node pair and the outgoing edge weight of each label node.

28. The apparatus according to claim 19, characterized in that, The device further includes: The page display module is used to obtain a display request for a media data search page, the display request carrying a target object; obtain various static display tags based on the display request, and obtain various initial media data based on the target object and the various static display tags; return the various initial media data and the various static display tags to the terminal corresponding to the display request, so as to display the various static display tags and the various initial media data on the media data search page through the terminal; The request acquisition module is also used to acquire the information recommendation request sent by the terminal through the media data search page. The information recommendation request carries the selected media data tag, which is determined from the various static display tags. The tag recommendation module is also used to return the target dynamic tag and the various recommended media data to the terminal corresponding to the information recommendation request, so as to display the target dynamic tag through the media data search page in the terminal, and update the various initial media data to the various recommended media data and display them.

29. The apparatus according to claim 28, characterized in that, The candidate tag determination module is also used to obtain the interest tags corresponding to the target object, and determine each candidate tag based on the interest tags and the tags associated with each recommended media data.

30. The apparatus according to claim 19, characterized in that, The device further includes: The dynamic tag selection module is used to obtain a target information recommendation request, which carries the selected media data tags and selected dynamic tags, the selected dynamic tags being determined from the target dynamic tags; based on the target information recommendation request, it obtains each target recommended media data associated with the selected dynamic tags from each recommended media data according to the preset association relationship; based on the selected media data tags and selected dynamic tags, it determines each target candidate tag, calculates the target tag recommendation degree corresponding to each target candidate tag, and determines each dynamic tag to be displayed from each target candidate tag based on the target tag recommendation degree; and returns each target recommended media data and each dynamic tag to be displayed to the terminal corresponding to the information recommendation request, so that the target recommended media data and each dynamic tag to be displayed can be displayed through the terminal.

31. An information display device, characterized in that, The device includes: The static tag display module is used to display various static tags on the media data search page; A dynamic tag display module is used to display a target dynamic tag on the media data search page in response to a selection event of a target static display tag among the various static display tags. The target dynamic tag is determined by identifying candidate tags from the tags corresponding to the recommended media data associated with the target static display tag, calculating the tag recommendation score based on the recommended media data associated with the target static display tag and the candidate media datasets associated with each candidate tag, and determining the tag recommendation score from the candidate tags based on the tag recommendation score. The tag is used to characterize the attribute information of the media data. The tag recommendation score corresponding to each candidate tag is obtained based on the candidate total quality information and data total quality information corresponding to each candidate tag. The candidate total quality information is calculated by summing the candidate quality information corresponding to the candidate media data in the candidate media dataset, and the data total quality information is calculated by summing the data quality information corresponding to each recommended media data. The candidate media dataset is associated with each candidate tag.

32. The apparatus according to claim 31, characterized in that, The dynamic tag display module is also used to respond to the selection event of the target static display tag among the various static display tags, and display the sub-tags corresponding to the target static display tag in the media data search page; In response to the selection event of a target sub-tag among the various sub-tags, each sub-dynamic tag is displayed on the media data search page; each sub-dynamic tag is determined by identifying each target candidate tag from the tags corresponding to the various recommended media data that are commonly associated with the target static display tag and the target sub-tag, calculating the recommendation degree based on the candidate media datasets associated with the commonly associated recommended media data and the target candidate tags respectively, obtaining the target tag recommendation degree corresponding to each target candidate tag, and determining it from the various target candidate tags based on the target tag recommendation degree.

33. The apparatus according to claim 31, characterized in that, The device further includes: The media data display module is used to respond to the selection event of the target dynamic tag by displaying the target dynamic tag in the tag adjustment area of ​​the media data search page, displaying each dynamic tag to be displayed in the dynamic tag display area of ​​the media data search page, and displaying each target recommended media data associated with the target dynamic tag in the data display area of ​​the media data search page. The dynamic tags to be displayed are obtained by recommending information using the target static display tag and the target dynamic tag as selected media data tags. The target recommended media data associated with the target dynamic tag is determined from the recommended media data associated with the target static display tag.

34. The apparatus according to claim 33, characterized in that, The target dynamic tags include at least two; The media data display module is further configured to, in response to the selection event of the at least two target dynamic tags, display the at least two target dynamic tags in the tag adjustment area of ​​the media data search page, display each target dynamic tag to be displayed in the dynamic tag display area of ​​the media data search page, and display each target recommended media data commonly associated with the at least two target dynamic tags in the data display area of ​​the media data search page. The each target dynamic tag to be displayed is obtained by recommending information using the target static display tag and the at least two target dynamic tags as selected media data tags. The each target recommended media data commonly associated with the at least two target dynamic tags is determined from the recommended media data associated with the target static display tag.

35. The apparatus according to claim 33, characterized in that, The device further includes: The deletion module is used to respond to a deletion event of the target dynamic tag in the tag adjustment area, delete the target dynamic tag in the tag adjustment area, and display the various recommended media data associated with the target static display tag in the data display area.

36. The apparatus according to claim 33, characterized in that, The device further includes: The history display module is used to respond to the selection event of the historically selected static display tag in each of the static display tags, display each historical dynamic tag in the media data search page, and display the historical target dynamic tag in the tag adjustment area. The historical target dynamic tag and each of the historical dynamic tags are recorded when the historically selected static display tag is deselected.

37. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 18.

38. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 18.

39. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1 to 18.

Images