Multimedia content recommendation method and device, equipment, medium and program product

Abstract

The invention discloses a multimedia content recommendation method and device, equipment, a medium and a program product, and relates to the field of content recommendation, the method comprises the following steps: obtaining playing information of a first multimedia content in a first client, the playing information being used for indicating a content belonging column of the first multimedia content played in the first client, a first account is logged in the first client; determining a preference label of the first account based on the playing information, wherein the preference label is a preference of the first account for playing the multimedia content determined based on the playing information; obtaining a second multimedia content based on the preference tag, wherein the second multimedia content is different from the first multimedia content in content type; and recommending the second multimedia content to the second client, wherein the association relationship between the first client and the second client is established by logging in the first account. Data sources of account preference analysis can be increased, the recommendation effect of multimedia content is improved, and the man-machine interaction efficiency is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of content recommendation, and in particular to a multimedia content recommendation method, apparatus, device, medium and program product. Background Art

[0002] Most music software can recommend songs that users are interested in, meet their personalized needs for music, and improve their experience.

[0003] In the related art, the music style and preference of users when listening to music using music software are analyzed, and songs are selected from the library for recommendation.

[0004] However, the single data source makes the analysis results relatively limited, and the recommendation results are highly homogeneous and have low accuracy, which makes it impossible for users to locate the songs they are interested in in a timely manner when listening to songs using the software, and the human-computer interaction efficiency is low. Summary of the invention

[0005] The embodiments of the present application provide a multimedia content recommendation method, device, equipment, medium and program product, which can improve the recommendation effect of multimedia content and improve the efficiency of human-computer interaction. The technical solution is as follows:

[0006] On the one hand, a multimedia content recommendation method is provided, the method comprising:

[0007] Acquire playback information of a first multimedia content in a first client, where the playback information is used to indicate a column to which the first multimedia content played in the first client belongs, and a first account is logged in to the first client;

[0008] Determine a preference tag of the first account based on the playback information, wherein the preference tag is a preference of the first account for playing multimedia content determined based on the playback information;

[0009] acquiring second multimedia content based on the preference tag, where the second multimedia content is of a different content type from the first multimedia content;

[0010] The second multimedia content is recommended to a second client, wherein the second client is used to display the second multimedia content on a recommendation page, and an association relationship between the first client and the second client is established by logging into the first account.

[0011] In another aspect, a multimedia content recommendation device is provided, the device comprising:

[0012] an acquisition module, configured to acquire playback information of a first multimedia content in a first client, wherein the playback information is used to indicate a column to which the first multimedia content played in the first client belongs, and a first account is logged in the first client;

[0013] a preference determination module, configured to determine a preference tag of the first account based on the playback information, wherein the preference tag is a preference of the first account for playing multimedia content determined based on the playback information;

[0014] a content acquisition module, configured to acquire second multimedia content based on the preference tag, wherein the second multimedia content is of a different content type from the first multimedia content;

[0015] The content recommendation module is used to recommend the second multimedia content to the second client, wherein the second client is used to display the second multimedia content on a recommendation page, and an association relationship between the first client and the second client is established by logging into the first account.

[0016] In an optional embodiment, the preference determination module is further used to determine a first production list of people who participated in the production of the column to which the content belongs based on the playback information, the first production list including the name of the producer of the first multimedia content; determine a first content tag of the first multimedia content based on the playback information, the first content tag being used to indicate a theme of the first multimedia content; and determine at least one of the first production list and the first content tag as a preference tag of the first account.

[0017] In an optional embodiment, the content acquisition module is further used to acquire a candidate content resource pool, wherein the candidate content resource pool includes multiple candidate multimedia content marked with content tags, and the content tags are used to indicate the themes of the candidate multimedia content; when the preference tag includes the first content tag, in response to the existence of a content tag of a candidate multimedia content matching the first content tag, at least one candidate multimedia content matching the first content tag is determined as the second multimedia content.

[0018] In an optional embodiment, the content acquisition module is further used to, when performing a matching analysis on the i-th candidate multimedia content in the candidate content resource pool, determine the i-th candidate multimedia content as the second multimedia content in response to a similarity between the first content tag and a content tag corresponding to the i-th candidate multimedia content reaching a first preset threshold; i is a positive integer.

[0019] In an optional embodiment, the content acquisition module is further used to, when performing a matching analysis on the jth candidate multimedia content in the candidate content resource pool, determine the jth candidate multimedia content as the second multimedia content in response to the similarity between the first content tag and the content tag corresponding to the jth candidate multimedia content not reaching the first preset threshold, and the similarity between the content tag corresponding to the i-th candidate multimedia content and the content tag corresponding to the j-th candidate multimedia content reaching the second preset threshold; j is a positive integer, and i and j are different.

[0020] In an optional embodiment, the content acquisition module is further used to obtain a candidate content resource pool, the candidate content resource pool includes multiple candidate multimedia content marked with production lists, and the production list includes the names of producers involved in the production of the candidate multimedia content; when the preference tag includes the first production list, in response to the existence of an overlap between the production list of the candidate multimedia content and the first production list that meets a preset list overlap requirement, at least one candidate multimedia content that meets the list overlap requirement is determined as the second multimedia content.

[0021] In an optional embodiment, the playback information includes a plurality of sub-information, the kth sub-information includes the kth playback timestamp information and the column information to which the kth content belongs, and k is a positive integer;

[0022] The acquisition module is also used to receive the kth playback image and the kth playback channel information sent by the first client when acquiring the kth sub-information, wherein the kth playback image is an image captured when the first client plays the first multimedia content, and after capturing the kth playback image, the first client identifies the kth playback image and determines the kth playback channel information; determines the kth playback timestamp information of the first multimedia content based on the kth playback image; determines the column information to which the kth content belongs based on the kth playback channel information and the kth playback timestamp information, and the kth content column information is used to indicate the column to which the content of the first multimedia content played by the first client in the kth sub-information belongs.

[0023] In an optional embodiment, the preference determination module is also used to, in response to the number of sub-information corresponding to the playback information reaching a preset quantity threshold, count the multiple sub-information and determine a target playback channel among the multiple sub-information that meets the preset playback time requirement; wherein the ratio of the playback time of the first multimedia content played through the target playback channel on the first client to the total playback time of the first multimedia content played on the first client meets the preset ratio requirement; obtain a list of columns played by the target playback channel in a historical time period, the list of columns containing at least one candidate column marked with a column label; determine a column among the at least one candidate column whose ratings meet the preset ratings requirement as a target column; and determine that the preference label of the first account contains a column label corresponding to the target column.

[0024] In an optional embodiment, the content acquisition module is further used to acquire a candidate content resource pool, the candidate content resource pool including a plurality of candidate multimedia contents marked with content tags, the content tags being used to indicate the themes of the candidate multimedia contents; when the preference tag includes a column tag corresponding to the target column, in response to the existence of a content tag of a candidate multimedia content matching the column tag, at least one candidate multimedia content matching the column tag is determined as the second multimedia content.

[0025] In an optional embodiment, the content acquisition module is further used to determine the pause timestamp of the first multimedia content when the playback information indicates that the first multimedia content is paused during playback; determine the synchronous playback content of the first account based on the playback information, wherein the first multimedia content includes the synchronous playback content; determine the start timestamp of playing the synchronous playback content based on the pause timestamp, and the start timestamp refers to the position in the progress bar of the synchronous playback content;

[0026] The content recommendation module is further used to recommend the synchronously played content to the second client, wherein the second client user displays the synchronously played content on a recommendation page, and the second client plays the synchronously played content starting from the start and end timestamps.

[0027] In an optional embodiment, the acquisition module is further used to acquire live broadcast information of the first client, where the live broadcast information is used to describe the third multimedia content broadcast live by the first client;

[0028] The preference determination module is further used to determine a live broadcast preference tag of the first account based on the live broadcast information, wherein the live broadcast preference tag is used to indicate the preference of the first account for watching multimedia content in the live broadcast;

[0029] The content acquisition module is further used to acquire fourth multimedia content based on the live broadcast preference tag;

[0030] The content recommendation module is further configured to recommend at least one of the fourth multimedia content and the third multimedia content to the second client.

[0031] On the other hand, a computer device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the multimedia content recommendation method as described in any of the above-mentioned embodiments of the present application.

[0032] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction, at least one program, a code set or an instruction set is stored in the storage medium, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement a multimedia content recommendation method as described in any of the above-mentioned embodiments of the present application.

[0033] On the other hand, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs any of the multimedia content recommendation methods described in the above embodiments.

[0034] The beneficial effects brought by the technical solution provided by the embodiment of the present application include at least:

[0035] By associating different clients with the same account, the preference of the first account for playing multimedia content is obtained based on the playback information of the multimedia content on one of the clients, which is used as the basis for recommending multimedia content to another client. Compared with the related art, the method of determining account preferences for content recommendation only through the historical records of the same client can increase the diversity of information sources, more accurately analyze the playback preferences of accounts, make precise recommendations, effectively alleviate the homogeneity problem in recommending multimedia content, help accounts expand or discover potential content of interest, and improve human-computer interaction efficiency. The multimedia content involved in analyzing account preferences is different from the type of multimedia content to be recommended, which can enrich the types of recommended content and improve the content recommendation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0037] Figure 1 is a schematic diagram of a multimedia content recommendation system provided by an exemplary embodiment of the present application;

[0038] Figure 2 is a flowchart of a multimedia content recommendation method provided by an exemplary embodiment of the present application;

[0039] Figure 3 is a schematic diagram of a recommendation page of a second client provided by an exemplary embodiment of the present application;

[0040] Figure 4 is a flowchart of a multimedia content recommendation method provided by another exemplary embodiment of the present application;

[0041] Figure 5 is a flowchart of a multimedia content recommendation method provided by another exemplary embodiment of the present application;

[0042] Figure 6 is a structural block diagram of a multimedia content recommendation device provided by an exemplary embodiment of the present application;

[0043] Figure 7 It is a structural block diagram of a computer device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0045] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0046] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0047] It should be noted that the information and data 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 relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0048] It should be understood that, although the terms first, second, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first parameter may also be referred to as the second parameter, and similarly, the second parameter may also be referred to as the first parameter. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0049] Most music applications have the function of recommending songs that users may like. Such customized services can meet users' personalized pursuit of music and enhance their user experience.

[0050] In the related art, the application determines the user's preferred music type and style by deeply analyzing the user's listening history on the application, and selects songs that meet the user's preferences from the music library for recommendation.

[0051] However, there are limitations to relying on a single data source for analysis. The recommended content is highly homogenized and lacks diversity, which affects the effectiveness and accuracy of recommendations. Users cannot find the songs they are interested in on the application recommendation page in a timely manner, and the human-computer interaction efficiency is low.

[0052] The present application provides a multimedia content recommendation method, which associates different clients with the same account, obtains the preference of the first account for playing multimedia content based on the playback information of the multimedia content on one of the clients, and serves as a basis for recommending multimedia content to another client. The multimedia content involved in analyzing the account preference is different from the type of multimedia content to be recommended, which can enrich the types of recommended content and improve the content recommendation effect and human-computer interaction efficiency.

[0053] The multimedia content recommendation system involved in the embodiments of the present application is described. For illustration, please refer to Figure 1 , Figure 1is a schematic diagram of a multimedia content recommendation system provided by an exemplary embodiment of the present application.

[0054] The multimedia content recommendation system includes a first client 110 , a second client 120 and a server (cloud) 130 , and a communication connection 140 exists among the first client 110 , the second client 120 and the server 130 .

[0055] Among them, the first client 110 is an application installed in the first terminal, and the second client 120 is an application installed in the second terminal. The first client 110 is logged in to the first account, and the first client 110 and the second client 120 are suggested to have an association relationship because they have logged in to the first account together.

[0056] It is worth noting that the first client 110 and the second client 120 can be applications installed in the same terminal device or in different terminal devices. This application takes the example of the first client 110 and the second client 120 being installed in different terminal devices.

[0057] The first client 110 and the second client 120 are multimedia content playback software, and the first client 110 and the second client 120 are mutually related applications. For example, the first client 110 and the second client 120 are different applications developed by the same enterprise.

[0058] When the first client 110 plays the first multimedia content, the first client collects the playback image at a preset frequency (e.g., once every 30 seconds), identifies the playback image, identifies the playback channel currently playing the first multimedia content, and sends the playback channel and the playback image as part of the playback information to the server 130. The server 130 determines the column to which the first multimedia content belongs based on the playback channel and the playback image, and obtains complete playback information.

[0059] The server 130 determines the preference of the first account when playing multimedia content based on the playback information, determines the preference tag of the first account, determines the second multimedia content from the resource pool based on the preference tag, and recommends it to the second client 120 .

[0060] After receiving the recommendation information, the second client 120 displays the second multimedia content on the recommendation page of the second client 120. For example, the second multimedia content is displayed in the form of a recommendation list, and the second multimedia content includes multimedia content 1, multimedia content 2, multimedia content 3, and multimedia content 4.

[0061] The first terminal and the second terminal may be terminal devices in various forms such as mobile phones, tablet computers, desktop computers, portable laptop computers, smart TVs, vehicle-mounted terminals, smart home devices, etc., and the embodiments of the present application are not limited to this.

[0062] It is worth noting that the above-mentioned servers can be independent physical servers, or they can be server clusters or distributed systems composed of multiple physical servers. They can also be cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), as well as big data and artificial intelligence platforms.

[0063] Among them, cloud technology refers to a hosting technology that unifies hardware, software, network and other resources in a wide area network or local area network to realize data computing, storage, processing and sharing. Cloud technology is a general term for network technology, information technology, integration technology, management platform technology, application technology, etc. based on the cloud computing business model. It can form a resource pool, which is used on demand and flexible and convenient. Cloud computing technology will become an important support. The background services of the technical network system require a large amount of computing and storage resources, such as video websites, picture websites and more portal websites. With the high development and application of the Internet industry, in the future, each item may have its own identification mark, and all need to be transmitted to the background system for logical processing. Data of different levels will be processed separately. All kinds of industry data require strong system backing support, which can only be achieved through cloud computing.

[0064] In some embodiments, the above server can also be implemented as a node in a blockchain system.

[0065] In combination with the above-mentioned noun introduction and application scenarios, the multimedia content recommendation method provided by the present application is described. The method can be executed by a server or a terminal, or can be executed by a server and a terminal together. In the embodiment of the present application, the method is described by taking the execution of the server as an example. Figure 2 As shown, Figure 2 FIG. 1 is a flowchart of a multimedia content recommendation method provided by an exemplary embodiment of the present application. The method includes the following steps.

[0066] Step 210: Acquire the playback information of the first multimedia content in the first client.

[0067] The play information is used to indicate the column to which the content of the first multimedia content played in the first client belongs, and the first account is logged in the first client.

[0068] Optionally, the types of the first multimedia content include but are not limited to: (1) text content: ordinary text, e-books, electronic documents, etc.; (2) image content: static pictures, dynamic pictures, such as GIF (Graphics Interchange Format), etc.; (3) audio content: music, radio dramas, audio books, recordings, sound effects, etc.; (4) video content: movies, TV programs, online videos, live videos, etc.; (5) animation content: two-dimensional animation, three-dimensional animation, etc.; (6) interactive content: online games, virtual reality games, online teaching content, etc.; (7) social media content: social media posts, pictures, videos, comments and chat messages posted on social media websites or applications, etc.; (8) presentation content, etc.

[0069] Exemplarily, the first client is an application installed in the smart TV terminal and capable of playing multimedia content, and the first client is logged in with a first account. The smart TV terminal runs the first client, and the first client is used to play the first multimedia content.

[0070] Among them, the first client can be an application installed in any terminal device, and the types of terminal devices include but are not limited to: (1) smart TV terminal; (2) broadcasting device terminal; (3) vehicle terminal: vehicle host, or other terminals installed inside the vehicle that can be used to play multimedia content; (4) desktop computer terminal; (5) tablet computer terminal; (6) smart watch terminal; (7) smart phone terminal, etc.

[0071] The types of the first client include, but are not limited to: (1) video playback application; (2) music playback application; (3) live broadcast application; (4) e-book reading application; (5) radio drama playback application, etc.

[0072] This embodiment is described by taking the example that the first client is a video playback application installed in a smart TV.

[0073] When the first multimedia content is played through the first client, the application program will collect a playing image, where the playing image is an image obtained by taking a screenshot of the screen of the first client.

[0074] The playback information is transmitted to the server side through the first client side. In some embodiments, the server side includes an analysis module that can analyze the playback image collected by the application program to obtain the playback information.

[0075] Exemplarily, the first multimedia content is a TV program content played through a first channel. When the first client plays the TV program content, it captures the playback image based on a preset frequency, for example, capturing a playback image every 30 seconds.

[0076] The collected playback image is identified to identify the playback channel information of the first multimedia content. The first client sends the playback channel information and the playback image to the server. The server analyzes the timestamp corresponding to the playback image based on the playback channel information and the playback image to determine the column to which the content of the first multimedia content belongs.

[0077] For example, the first multimedia content is a music program, and the playback image includes a picture of multiple singers singing a first song and the lyrics of the first song, then the content of the first multimedia content belongs to the music column.

[0078] Step 220: Determine a preference tag of the first account based on the playback information.

[0079] The preference tag is a preference of the first account for playing multimedia content determined based on the playback information.

[0080] Optionally, a first production list of the column to which the content belongs that participates in production is determined based on the playback information, and the first production list includes the name of the producer of the first multimedia content.

[0081] For example, the first multimedia content is a music program, and the first production list includes the names of the producer, director, music director, composer, host, and singer who participated in the production of the program.

[0082] A first content tag of the first multimedia content is determined based on the playback information, where the first content tag is used to indicate a theme of the first multimedia content.

[0083] For example, the first multimedia content is a music program, and the first content tags corresponding to the singers participating in the program are as follows: song, music, variety show, language, hit song, and the name of the singer participating in the program.

[0084] The first content label can be obtained in the following ways: (1) after analyzing the first multimedia content, the staff manually labels the first multimedia content; (2) according to the name of the first multimedia content, the keywords are counted in the articles, blogs, and comments that publicly discuss or mention the first multimedia content on the social network, and the keywords are used as the first content label, wherein the keywords refer to words whose appearance frequency meets the preset frequency requirements; (3) the information disclosed by the first multimedia content or the file obtained after downloading the first multimedia content is input into the pre-trained model, and the pre-trained model analyzes the input data and outputs the first content label, wherein the type of the pre-trained model is determined according to the type of the first multimedia content. For example, if the first multimedia content is a language program, a natural language model is used as the pre-trained model.

[0085] At least one of the first production list and the first content tag is determined as a preference tag of the first account.

[0086] The preference tags in the example can reflect the first account's preference when playing music programs, such as preferring music programs produced by A, preferring songs sung by singer B, etc.

[0087] Step 230: Acquire second multimedia content based on the preference tag.

[0088] The types of the second multimedia content include but are not limited to: (1) text content; (2) image content;

[0089] (3) Audio content; (4) Video content; (5) Animation content; (6) Interactive content; (7) Social media content; (8) Presentation content, etc.

[0090] The second multimedia content is of a different content type from the first multimedia content. For example, the first multimedia content is a video content: a TV program; and the second multimedia content is an audio content: a song.

[0091] By using different types of multimedia content to determine preference tags and recommend multimedia content to the first account, the first account's playback preferences can be analyzed more comprehensively. While accurately recommending content to the first account, the problem of content homogeneity caused by content recommendation based only on the same type of multimedia content can be effectively solved. This enables the first account to quickly locate content of interest and discover potentially attractive content, thereby improving the effectiveness of content recommendation and the efficiency of human-computer interaction.

[0092] Optionally, a candidate content resource pool is obtained, wherein the candidate content resource pool includes a plurality of candidate multimedia contents marked with content tags, where the content tags are used to indicate themes of the candidate multimedia contents.

[0093] The content tags of the candidate multimedia content may be determined by at least one of the following methods: (1) manual labeling; (2) keyword extraction; (3) model analysis.

[0094] The content type of the candidate multimedia content is different from the content type of the first multimedia content. The content tag of the candidate multimedia content and the first content tag may be determined in the same manner or in a different manner.

[0095] When the preference tag includes the first content tag, in response to the presence of a content tag of the candidate multimedia content matching the first content tag, at least one candidate multimedia content matching the first content tag is determined as the second multimedia content.

[0096] Exemplarily, when performing matching analysis on the i-th candidate multimedia content in the candidate content resource pool, in response to the similarity between the first content tag and the content tag corresponding to the i-th candidate multimedia content reaching a first preset threshold, the i-th candidate multimedia content is determined as the second multimedia content; i is a positive integer.

[0097] Wherein, taking the content tag of the i-th candidate multimedia content as a reference, first determine whether there is an overlapping part between the first content tag and the content tag of the i-th candidate multimedia content, and determine the similarity of the overlapping part as 1. If there is a non-overlapping part, the content tag of the candidate multimedia content in the non-overlapping part includes at least one tag, and the similarity between the m-th content tag and the tag belonging to the first content tag in the non-overlapping part is calculated, and the one with the highest similarity value is taken as the similarity of the m-th content tag in the content tags of the i-th candidate multimedia content.

[0098] A weighted operation is performed on each content tag of the i-th candidate multimedia content to obtain a similarity between the first content tag and the content tag corresponding to the i-th candidate multimedia content.

[0099] For example, the first content tag is: "Music Variety Show", "Golden Song", "Name of Singer A", "Folk Song"; the content tags of the i-th candidate multimedia content include: [Song], [Name of Singer A], [Folk Song].

[0100] Overlapping parts: singer A’s name and folk songs; non-overlapping parts: music variety shows, golden songs, and songs.

[0101] The similarities between [name of singer A] and [folk song] and the first content tag are both 100%.

[0102] In the non-overlapping part, the similarity between [song] and "Music Variety" is 50%, the similarity between [song] and "Golden Song" is 90%, and the similarity between [song] and the first content tag is 90%.

[0103] A weighted operation is performed on each content tag: (100%+100%+90% / 3)=96.67%, the first preset threshold is 90%, the similarity between the first content tag and the content tag corresponding to the i-th candidate multimedia content reaches the first preset threshold, and the i-th candidate multimedia content is determined as the second multimedia content.

[0104] It is worth noting that the number of content tags of each candidate multimedia content and the number of the first content tag can be 1, in which case the similarity between the two tags can be directly calculated. The similarity can be determined in any manner, for example, by extracting features from the content tags in text form to obtain feature vectors, and calculating the vector distance or cosine similarity between the feature vectors as the similarity between the first content tag and the content tag of the candidate multimedia content.

[0105] In some embodiments, when performing a matching analysis on the jth candidate multimedia content in the candidate content resource pool, in response to the similarity between the first content tag and the content tag corresponding to the jth candidate multimedia content not reaching a first preset threshold, and the similarity between the content tag corresponding to the i-th candidate multimedia content and the content tag corresponding to the j-th candidate multimedia content reaches a second preset threshold, the j-th candidate multimedia content is determined as the second multimedia content; j is a positive integer, and i and j are different.

[0106] For example, the similarity between the content tag of the fifth candidate multimedia content and the first content tag is 88%, which does not reach the first preset threshold of 90%; the similarity between the content tag of the second candidate multimedia content and the first content tag is 98%, which reaches the first preset threshold of 90%; the similarity between the content tag of the fifth candidate multimedia content and the content tag of the second candidate multimedia content is 93%, which reaches the second preset threshold of 92%, and the fifth candidate multimedia content is determined as the second multimedia content.

[0107] Optionally, a candidate content resource pool is obtained, wherein the candidate content resource pool includes a plurality of candidate multimedia contents marked with production lists, and the production lists include names of producers who participated in producing the candidate multimedia contents.

[0108] When the preference tag includes the first production list, in response to the overlap between the production list of the candidate multimedia content and the first production list meeting the preset list overlap requirement, at least one candidate multimedia content meeting the list overlap requirement is determined as the second multimedia content.

[0109] Exemplarily, the list overlap requirement means that the ratio of the number of names in the production list of the candidate multimedia content that overlap with the first production list to the total number of names in the production list of the candidate multimedia content reaches a first ratio threshold.

[0110] For example, the first proportion threshold is 50%, the first production list contains 30 producer names, the production list of the candidate multimedia content contains 25 producer names, and the number of overlapping names is 15, then the proportion is 15 / 25=60%, reaching the first proportion threshold and meeting the preset list overlap requirement, and the candidate multimedia content is determined as the second multimedia content.

[0111] Step 240: recommend the second multimedia content to the second client.

[0112] The second client is used to display the second multimedia content on the recommendation page, and the first client and the second client establish an association relationship between the clients by logging into the first account.

[0113] Exemplarily, the second client is an application installed in a smart phone that can play multimedia content, and the second client can log in to the first account.

[0114] The second client may be an application installed in any terminal device, and the types of terminal devices include but are not limited to: (1) smart TV terminal; (2) broadcasting device terminal; (3) vehicle terminal: vehicle host, or other terminals installed inside the vehicle that can be used to play multimedia content; (4) desktop computer terminal; (5) tablet computer terminal; (6) smart watch terminal; (7) smart phone terminal, etc.

[0115] The types of the second client include, but are not limited to: (1) video playback application; (2) music playback application; (3) live broadcast application; (4) e-book reading application; (5) radio drama playback application, etc.

[0116] This embodiment is described by taking the second client being a video playback application installed in a smart phone as an example. The first client and the second client have at least one of the following associations.

[0117] (1) Different applications developed by the same enterprise / individual can communicate and transfer data between the two applications, and the first account can log in to the two applications at the same time; (2) Different applications developed by different enterprises / individuals can communicate and transfer data between the two applications after mutual authorization, and a binding relationship or association relationship is established between the accounts logged in to the two applications.

[0118] Optionally, after determining the second multimedia content to be recommended, the server sends index information corresponding to the second multimedia content to the second client, where the index information is used to instruct the second client to obtain the second multimedia content from the candidate content resource pool.

[0119] Exemplarily, the candidate content resource pool includes multiple types of multimedia content, each type of multimedia content has its own classification number and a sequence number under the category entry, and the index information includes the classification number and the sequence number of the second multimedia content.

[0120] For example, the candidate content resource pool includes 7 types of multimedia content, each type of multimedia content has a classification number of A to G, and the number of multimedia content under each category entry is 1000, 20000, 15000, 1200, 1900, 30000, and 9000. The index information indicates that the classification sequence of the second multimedia content is A and the sequence number is 400, indicating that the second multimedia content is the multimedia content ranked 400 in the A-type multimedia content.

[0121] The second client obtains the second multimedia content based on the index information, and displays introduction information of the second multimedia content on the recommendation page to attract users to play the content.

[0122] Schematically, the recommendation page of the second client is as follows Figure 3 As shown, the recommendation page includes a recommendation list 310, which includes multiple second multimedia contents to be recommended. Each multimedia content has a corresponding name, such as: multimedia content 1, multimedia content 2, multimedia content 3, multimedia content 4.

[0123] Exemplarily, multimedia content 1, multimedia content 2, multimedia content 3, and multimedia content 4 are all audio content.

[0124] In summary, the multimedia content recommendation method provided by the present application associates different clients with the same account, obtains the preference of the first account for playing multimedia content based on the playback information of the multimedia content on one of the clients, and serves as the basis for recommending multimedia content to another client. Compared with the related art, the method of determining account preferences for content recommendation only through the historical records of the same client can increase the diversity of information sources, more accurately analyze the playback preferences of accounts, make precise recommendations, effectively alleviate the homogeneity problem when recommending multimedia content, and help accounts expand or discover potential content of interest. The multimedia content involved in analyzing account preferences improves the efficiency of human-computer interaction, and is different from the type of multimedia content to be recommended, which can enrich the types of recommended content and improve the content recommendation effect.

[0125] Figure 4 is a flowchart of a multimedia content recommendation method provided by another exemplary embodiment of the present application, the method comprising the following steps.

[0126] Step 410: Acquire the playback information of the first multimedia content in the first client.

[0127] The play information is used to indicate the column to which the content of the first multimedia content played in the first client belongs, and the first account is logged in the first client.

[0128] Optionally, in this embodiment, the first client is a video playback application installed in a smart TV as an example for description.

[0129] The play information includes a plurality of sub-information, and the k-th sub-information includes the k-th play timestamp information and the column information to which the k-th content belongs, where k is a positive integer.

[0130] When the first client plays the first multimedia content, the first client collects information based on a preset frequency and sends it to the server. The server analyzes the information each time it receives the information sent by the first client to obtain a sub-information. For example, the first client sends information to the server once every 30 seconds. After sending information to the server for the first time, the server receives the first information and analyzes it to obtain the first sub-information.

[0131] Exemplarily, in the case of acquiring the kth sub-information, the kth playback image and the kth playback channel information sent by the first client are received.

[0132] The kth playback image is an image captured when the first client plays the first multimedia content. After capturing the kth playback image, the first client identifies the kth playback image and determines the kth playback channel information.

[0133] That is, the first multimedia content played by the first client is composed of at least one multimedia content, and the user may change the channel or the type of multimedia content when watching the multimedia content played by the first client.

[0134] For example, the first client plays the first multimedia content in the time period from 13:00 to 14:00, plays multimedia content A (a documentary program) in the time period from 13:00 to 13:45, and plays multimedia content B (a variety show) in the time period from 13:45 to 14:00. The playback channel information corresponding to multimedia content A is the first channel, and the playback channel information corresponding to multimedia content B is the second channel.

[0135] The kth playback timestamp information of the first multimedia content is determined based on the kth playback image, where the playback timestamp information is used to indicate the timestamp of capturing the playback image, and the timestamp information can reflect the progress of the first multimedia content when it is broadcast on the channel.

[0136] The column information of the kth content is determined based on the kth playing channel information and the kth playing timestamp information, and the column information of the kth content is used to indicate the column to which the first multimedia content played by the first client in the kth sub-information belongs.

[0137] For example, the k-th playback timestamp information is 13:00 on October 1, 2020. At this time, the playback channel information corresponding to the first multimedia content is the first channel. The first channel broadcasts a music variety show between 12:00 and 14:00 on October 1, 2020. The column to which the first multimedia content belongs is "Music Variety Show". The k-th playback timestamp indicates that the playback progress of the variety show is 1 hour (1 / 2).

[0138] Step 420, in response to the number of sub-information corresponding to the play information reaching a preset number threshold, statistics are collected on the multiple pieces of sub-information to determine a target play channel that meets the preset play duration requirement among the multiple pieces of sub-information.

[0139] The ratio of the playing time of the first multimedia content played on the first client through the target playing channel to the total playing time of the first multimedia content played on the first client meets the preset ratio requirement.

[0140] Exemplarily, the preset quantity threshold of the sub-information is 100, and the preset ratio requirement means that the ratio of the play time of the target play channel to the total play time is the highest among all play channels.

[0141] The playback information obtained by the server side includes 150 sub-information, among which the playback channel information of the first multimedia content in the 1st to 50th sub-information is the first channel, the playback channel information of the first multimedia content in the 51st to 90th sub-information is the second channel, and the playback channel information of the first multimedia content in the 91st to 150th sub-information is the third channel.

[0142] Since the duration interval of each sub-information is the same, it can be calculated that the ratio of the playback time of the first channel to the total playback time is 50 / 150=1 / 3, the ratio of the playback time of the second channel to the total playback time is 40 / 150=4 / 15, and the ratio of the playback time of the third channel to the total playback time is 60 / 150=2 / 5.

[0143] The ratio of the playback time of the third channel to the total playback time is the highest, so the third channel is determined as the target playback channel.

[0144] Step 430, obtaining a list of programs played by the target broadcast channel in a historical time period.

[0145] The column list includes at least one candidate column marked with a column tag.

[0146] For example, the target broadcast channel is a local TV station channel, and the candidate columns broadcast by the target broadcast channel in the recent day include: TV series column, music variety column, cartoon column and news column.

[0147] The column label of the TV series column includes at least one of the TV series name, the name of the leading actor, the name of the director, the theme of the TV series, and the total duration of the TV series.

[0148] The column label of the music variety show column includes at least one of the variety show name, the name of the singer, the name of the arranger, the singing repertoire, and the total duration of the variety show.

[0149] The column label of the cartoon column includes at least one of the cartoon name, the voice actor name, the director name, the cartoon theme, and the total duration of the cartoon.

[0150] The column label of the news column includes at least one of the host name, the classification of the reported event, and the daily broadcast duration of the news.

[0151] The column labels of the candidate columns may be obtained by at least one of the following methods: (1) manual labeling; (2) keyword extraction; (3) model analysis.

[0152] Step 440: determine a column whose ratings meet the preset ratings requirement among at least one candidate column as a target column, and determine that the preference label of the first account includes a column label corresponding to the target column.

[0153] Optionally, the preset audience requirement means that the average audience rating of the column in the past month is the highest among all candidate columns.

[0154] For example, if the target broadcast channel has only one candidate column, the label of the candidate column is used as the preference label of the first account.

[0155] For example, the target broadcast channel includes 4 candidate programs: candidate program 1, candidate program 2, candidate program 3, and candidate program 4. The average audience ratings of each candidate program are 20%, 23%, 15%, and 10%, respectively.

[0156] Among them, the average ratings of candidate column 2 are the highest at 23%, and candidate column 2 is determined as the target column, and the column label corresponding to candidate column 2 is determined to be included in the preference label of the first account.

[0157] Step 450: Acquire a candidate content resource pool.

[0158] The candidate content resource pool includes a plurality of candidate multimedia contents marked with content tags, where the content tags are used to indicate the subject of the candidate multimedia contents.

[0159] The content tags of the candidate multimedia content may be determined by at least one of the following methods: (1) manual labeling; (2) keyword extraction; (3) model analysis.

[0160] Step 460: when the preference tag includes a column tag corresponding to the target column, in response to the existence of a content tag of the candidate multimedia content matching the column tag, determine at least one candidate multimedia content matching the column tag as the second multimedia content.

[0161] Exemplarily, when performing matching analysis on the i-th candidate multimedia content in the candidate content resource pool, in response to the similarity between the column tag and the content tag corresponding to the i-th candidate multimedia content reaching a first preset threshold, the i-th candidate multimedia content is determined as the second multimedia content; i is a positive integer.

[0162] Wherein, taking the content tag of the i-th candidate multimedia content as a reference, first determine whether there is an overlapping part between the column tag and the content tag of the i-th candidate multimedia content, and determine the similarity of the overlapping part as 1. If there is a non-overlapping part, the content tag of the candidate multimedia content in the non-overlapping part includes at least one tag, and the similarity between the m-th content tag and the tag belonging to the first content tag in the non-overlapping part is calculated respectively, and the one with the highest similarity value is taken as the similarity of the m-th content tag in the content tags of the i-th candidate multimedia content.

[0163] A weighted operation is performed on each content tag of the i-th candidate multimedia content to obtain a similarity between the first content tag and the content tag corresponding to the i-th candidate multimedia content.

[0164] Step 470: recommend the second multimedia content to the second client.

[0165] The second client is used to display the second multimedia content on the recommendation page, and the first client and the second client establish an association relationship between the clients by logging into the first account.

[0166] Exemplarily, the second client is an application installed in a smart phone that can play multimedia content, and the first account is logged in through the application.

[0167] For example, the second client is a music player application in a smart phone.

[0168] In summary, the multimedia content recommendation method provided by the present application obtains playback information (sub-information) once based on a preset frequency, can grasp the situation of the first client playing the first multimedia content in real time, and can improve the accuracy of the analysis results by performing preference analysis based on the column labels of the playback columns with the highest playback time ratio, obtain preference labels that are more in line with the playback preferences of the first account, and improve the effect of multimedia content recommendation based on preference labels and the efficiency of human-computer interaction. Using the labels of the first multimedia content that are different from the content type of the second multimedia content for analysis can avoid the problem of homogenization of recommendation results.

[0169] Figure 5 is a flowchart of a multimedia content recommendation method provided by another exemplary embodiment of the present application, the method comprising the following steps.

[0170] Step 510: Acquire the playback information of the first multimedia content in the first client.

[0171] The play information is used to indicate the column to which the content of the first multimedia content played in the first client belongs, and the first account is logged in the first client.

[0172] Step 520: When the playback information indicates that the first multimedia content is paused during playback, determine a pause timestamp of the first multimedia content.

[0173] The pause timestamp is used to indicate the moment when the first multimedia content is paused.

[0174] For example, the first client starts playing the first multimedia content in the first channel at 13:00:00 on January 1, 2020, and pauses at 13:09:30.

[0175] Then, the pause timestamp is January 1, 2020 13:09:30.

[0176] Step 530: Determine the synchronous playback content of the first account based on the playback information.

[0177] The first multimedia content includes synchronous playback content.

[0178] Exemplarily, the playback information indicates that the first multimedia content is a music variety show. At 13:09:30 on January 1, 2020, the first multimedia content includes a picture of singer A singing track B. Then the synchronized playback content of the first account is track B sung by singer A, that is, the synchronized playback content is the song content.

[0179] Step 540: Determine a start timestamp for playing the synchronously played content based on the pause timestamp.

[0180] The start timestamp refers to the position in the progress bar of the synchronously played content.

[0181] Exemplarily, the pause timestamp is 13:09:30 on January 1, 2020, and the synchronously played content is song content B in the first multimedia content broadcast by the first channel. The song content B starts broadcasting at 13:07:34, and its broadcast timestamp is 13:07:34 on January 1, 2020.

[0182] The start timestamp is determined based on the time difference between the pause timestamp and the broadcast timestamp. The time difference is 1 minute and 56 seconds, so the start timestamp of song content B is 1 minute and 56 seconds.

[0183] Step 550: recommending synchronous playback content to the second client.

[0184] The second client user displays the synchronous playback content on the recommendation page, and the second client plays the synchronous playback content starting from the start and end timestamps.

[0185] For example, the synchronous playback content is song content B, the song length of song content B is 3 minutes, and the starting timestamp is 1 minute and 56 seconds, which means that when the second client plays song content B, it starts playing from 1 minute and 56 seconds of song content B, realizing synchronous playback of multimedia content on different clients.

[0186] In some embodiments, the synchronously played content is the same as the first multimedia content, that is, when the first client pauses while playing the first multimedia content, the pause timestamp of the first multimedia content is obtained, and the first multimedia content is continued to be played from the pause timestamp on the second client to achieve synchronous playback of the multimedia content.

[0187] In some embodiments, if the first account watches the live broadcast in real time through the first client, the multimedia content played in the live broadcast can also be synchronously recommended to the second client, and the preference of the first account for watching the live broadcast content can be recorded in time.

[0188] Optionally, live broadcast information of the first client is obtained, where the live broadcast information is used to describe the third multimedia content that is broadcast live by the first client.

[0189] Exemplarily, a first account watches a second account's live broadcast room through a first client, and the live broadcast room plays third multimedia content. When the first client plays the third multimedia content in a live broadcast manner, the live broadcast information of the first client is obtained in real time.

[0190] For example, the third multimedia content played in the live broadcast room of the second account is song content, and the live broadcast information includes song information of the third multimedia content: the song title and the name of the singer of the song corresponding to at least one song.

[0191] A live broadcast preference tag of the first account is determined based on the live broadcast information, where the live broadcast preference tag is used to indicate the preference of the first account for watching multimedia content in the live broadcast.

[0192] Exemplarily, the live broadcast information indicates song information of at least one song, and a song tag corresponding to the song is obtained based on the song information. The song tag can be obtained in at least one of the following ways: (1) manual labeling; (2) keyword extraction; (3) model analysis.

[0193] For example, the third multimedia content includes three songs: Song A, Song B, and Song C. Based on the live broadcast information, the song tags corresponding to each song are determined to be: sad, inspirational, and growth.

[0194] The fourth multimedia content is acquired based on the live broadcast preference tag.

[0195] Exemplarily, a candidate content resource pool is obtained.

[0196] The candidate content resource pool includes a plurality of candidate multimedia contents marked with content tags, where the content tags are used to indicate the subject of the candidate multimedia contents.

[0197] In response to the existence of a content tag of the candidate multimedia content matching the song tag, at least one candidate multimedia content matching the song tag is determined as the fourth multimedia content.

[0198] At least one of the fourth multimedia content and the third multimedia content is recommended to the second client.

[0199] That is, the third multimedia content involved in the live broadcast can be recorded and synchronized to the second client, and the fourth multimedia content that the first account may be interested in can be determined based on the first account's preference when watching the live broadcast on the first client, and the fourth multimedia content can be recommended and displayed on the second client.

[0200] In summary, the multimedia content recommendation method provided in the present application can timely obtain the pause status of the first multimedia content, determine the starting timestamp when the multimedia content is synchronously played on different clients, and achieve the effect of synchronous and coherent playback of multimedia content across clients.

[0201] Figure 6 is a structural block diagram of a multimedia content recommendation device provided by an exemplary embodiment of the present application. Figure 6 As shown, the device includes the following parts.

[0202] The acquisition module 610 is used to acquire the playback information of the first multimedia content in the first client, where the playback information is used to indicate the column to which the content of the first multimedia content played in the first client belongs, and the first account is logged in the first client;

[0203] A preference determination module 620, configured to determine a preference tag of the first account based on the playback information, wherein the preference tag is a preference of the first account for playing multimedia content determined based on the playback information;

[0204] A content acquisition module 630, configured to acquire second multimedia content based on the preference tag, wherein the second multimedia content is of a different content type from the first multimedia content;

[0205] The content recommendation module 640 is used to recommend the second multimedia content to the second client, wherein the second client is used to display the second multimedia content on a recommendation page, and an association relationship between the first client and the second client is established by logging into the first account.

[0206] In an optional embodiment, the preference determination module 620 is further used to determine a first production list of people who participated in the production of the column to which the content belongs based on the playback information, the first production list including the name of the producer of the first multimedia content; determine a first content tag of the first multimedia content based on the playback information, the first content tag being used to indicate the theme of the first multimedia content; and determine at least one of the first production list and the first content tag as a preference tag of the first account.

[0207] In an optional embodiment, the content acquisition module 630 is further used to acquire a candidate content resource pool, wherein the candidate content resource pool includes multiple candidate multimedia content marked with content tags, and the content tags are used to indicate the topics of the candidate multimedia content; when the preference tag includes the first content tag, in response to the existence of a content tag of a candidate multimedia content matching the first content tag, at least one candidate multimedia content matching the first content tag is determined as the second multimedia content.

[0208] In an optional embodiment, the content acquisition module 630 is further used to, when performing a matching analysis on the i-th candidate multimedia content in the candidate content resource pool, determine the i-th candidate multimedia content as the second multimedia content in response to the similarity between the first content tag and the content tag corresponding to the i-th candidate multimedia content reaching a first preset threshold; i is a positive integer.

[0209] In an optional embodiment, the content acquisition module 630 is further used to, when performing a matching analysis on the jth candidate multimedia content in the candidate content resource pool, determine the jth candidate multimedia content as the second multimedia content in response to the similarity between the first content tag and the content tag corresponding to the jth candidate multimedia content not reaching the first preset threshold, and the similarity between the content tag corresponding to the i-th candidate multimedia content and the content tag corresponding to the j-th candidate multimedia content reaching the second preset threshold; j is a positive integer, and i and j are different.

[0210] In an optional embodiment, the content acquisition module 630 is further used to obtain a candidate content resource pool, wherein the candidate content resource pool includes multiple candidate multimedia content marked with a production list, and the production list includes the names of producers involved in the production of the candidate multimedia content; when the preference tag includes the first production list, in response to the existence of an overlap between the production list of the candidate multimedia content and the first production list that meets a preset list overlap requirement, at least one candidate multimedia content that meets the list overlap requirement is determined as the second multimedia content.

[0211] In an optional embodiment, the playback information includes a plurality of sub-information, the kth sub-information includes the kth playback timestamp information and the column information to which the kth content belongs, and k is a positive integer;

[0212] The acquisition module 610 is also used to receive the kth playback image and the kth playback channel information sent by the first client when the kth sub-information is acquired, wherein the kth playback image is an image captured when the first client plays the first multimedia content, and after capturing the kth playback image, the first client identifies the kth playback image and determines the kth playback channel information; determines the kth playback timestamp information of the first multimedia content based on the kth playback image; determines the column information to which the kth content belongs based on the kth playback channel information and the kth playback timestamp information, and the kth content column information is used to indicate the column to which the content of the first multimedia content played by the first client in the kth sub-information belongs.

[0213] In an optional embodiment, the preference determination module 620 is also used to, in response to the number of sub-information corresponding to the playback information reaching a preset quantity threshold, count the multiple sub-information and determine a target playback channel in the multiple sub-information that meets the preset playback time requirement; wherein the ratio of the playback time of the first multimedia content played through the target playback channel on the first client to the total playback time of the first multimedia content played on the first client meets the preset ratio requirement; obtain a list of columns played by the target playback channel in a historical time period, the list of columns containing at least one candidate column marked with a column label; determine the column whose ratings meet the preset ratings requirement among the at least one candidate column as the target column; and determine that the preference label of the first account contains the column label corresponding to the target column.

[0214] In an optional embodiment, the content acquisition module 630 is also used to obtain a candidate content resource pool, which includes multiple candidate multimedia content marked with content tags, and the content tags are used to indicate the themes of the candidate multimedia content; when the preference tag includes a column tag corresponding to the target column, in response to the existence of a content tag of a candidate multimedia content that matches the column tag, at least one candidate multimedia content that matches the column tag is determined as the second multimedia content.

[0215] In an optional embodiment, the content acquisition module 630 is further used to determine the pause timestamp of the first multimedia content when the playback information indicates that the first multimedia content is paused during playback; determine the synchronous playback content of the first account based on the playback information, wherein the first multimedia content includes the synchronous playback content; determine the start timestamp of playing the synchronous playback content based on the pause timestamp, and the start timestamp refers to the position in the progress bar of the synchronous playback content;

[0216] The content recommendation module 640 is further configured to recommend the synchronously played content to the second client, wherein the second client user displays the synchronously played content on a recommendation page, and the second client plays the synchronously played content starting from the start and end timestamps.

[0217] In an optional embodiment, the acquisition module 610 is further used to acquire live broadcast information of the first client, where the live broadcast information is used to describe the third multimedia content broadcast live by the first client;

[0218] The preference determination module 620 is further configured to determine a live broadcast preference tag of the first account based on the live broadcast information, wherein the live broadcast preference tag is used to indicate the preference of the first account for viewing multimedia content in the live broadcast;

[0219] The content acquisition module 630 is further configured to acquire fourth multimedia content based on the live broadcast preference tag;

[0220] The content recommendation module 640 is further configured to recommend at least one of the fourth multimedia content and the third multimedia content to the second client.

[0221] In summary, the multimedia content recommendation device provided by the present application associates different clients with the same account, obtains the preference of the first account for playing multimedia content based on the playback information of the multimedia content on one of the clients, and serves as the basis for recommending multimedia content to another client. Compared with the related art, the method of determining account preferences for content recommendation only through the historical records of the same client can increase the diversity of information sources, more accurately analyze the playback preferences of accounts, make precise recommendations, effectively alleviate the homogeneity problem in recommending multimedia content, help accounts expand or discover potential content of interest, and improve human-computer interaction efficiency. The multimedia content involved in analyzing account preferences is different from the type of multimedia content to be recommended, which can enrich the types of recommended content and improve the content recommendation effect.

[0222] It should be noted that the multimedia content recommendation device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the multimedia content recommendation device provided in the above embodiment and the multimedia content recommendation method embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0223] Figure 7 The block diagram of the structure of a computer device 700 provided by an exemplary embodiment of the present application is shown. The computer device 700 may be: a smart phone, a tablet computer, an MP3 player (Moving Picture Experts Group Audio Layer III, Moving Picture Experts Group Audio Layer 3), an MP4 (Moving Picture Experts Group Audio Layer IV, Moving Picture Experts Group Audio Layer 4) player, a laptop computer or a desktop computer. The computer device 700 may also be called a user device, a portable terminal, a laptop terminal, a desktop terminal or other names.

[0224] Typically, the computer device 700 includes a processor 701 and a memory 702 .

[0225] The processor 701 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 701 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 701 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 701 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 701 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0226] The memory 702 may include one or more computer-readable storage media, which may be non-transitory. The memory 702 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 702 is used to store at least one instruction, which is used to be executed by the processor 701 to implement the multimedia content recommendation method provided in the method embodiment of the present application.

[0227] In some embodiments, the computer device 700 further includes some other components 703, and the type and quantity of the other components 703 can be selected based on the functional requirements of the computer device 700. Those skilled in the art will appreciate that Figure 7 The structure shown in the figure does not constitute a limitation on the computer device 700, and the computer device 700 may include more or less components than shown in the figure, or combine some components, or adopt a different arrangement of components.

[0228] Optionally, the computer readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a solid state drive (SSD), or an optical disk. Among them, the random access memory may include a resistance random access memory (ReRAM) and a dynamic random access memory (DRAM). The serial numbers of the above embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0229] An embodiment of the present application also provides a computer device, which includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the multimedia content recommendation method as described in any of the above embodiments of the present application.

[0230] An embodiment of the present application also provides a computer-readable storage medium, in which at least one instruction, at least one program, a code set or an instruction set is stored. The at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement a multimedia content recommendation method as described in any of the above embodiments of the present application.

[0231] The embodiment of the present application also provides a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device performs any of the multimedia content recommendation methods described in the above embodiments.

[0232] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0233] The above description is only an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A multimedia content recommendation method, characterized in that: The method comprises: Acquire playback information of a first multimedia content in a first client, where the playback information is used to indicate a column to which the first multimedia content played in the first client belongs, and a first account is logged in to the first client; Determine a preference tag of the first account based on the playback information, wherein the preference tag is a preference of the first account for playing multimedia content determined based on the playback information; acquiring second multimedia content based on the preference tag, where the second multimedia content is of a different content type from the first multimedia content; The second multimedia content is recommended to a second client, wherein the second client is used to display the second multimedia content on a recommendation page, and an association relationship between the first client and the second client is established by logging into the first account.

2. The method according to claim 1, characterized in that The determining the preference tag of the first account based on the playback information includes: Determine a first production list of people who participated in producing the column to which the content belongs based on the playback information, wherein the first production list includes the name of a producer of the first multimedia content; determining a first content tag of the first multimedia content based on the playback information, where the first content tag is used to indicate a theme of the first multimedia content; At least one of the first production list and the first content tag is determined as a preference tag of the first account.

3. The method according to claim 2, characterized in that The acquiring the second multimedia content based on the preference tag includes: Acquire a candidate content resource pool, wherein the candidate content resource pool includes a plurality of candidate multimedia contents marked with content tags, wherein the content tags are used to indicate the subject of the candidate multimedia contents; In the case where the preference tag includes the first content tag, in response to the presence of a candidate multimedia content having a content tag matching the first content tag, at least one candidate multimedia content matching the first content tag is determined as the second multimedia content.

4. The method according to claim 3, characterized in that In response to the content tag of the candidate multimedia content matching the first content tag, determining at least one candidate multimedia content matching the first content tag as the second multimedia content includes: When performing matching analysis on the i-th candidate multimedia content in the candidate content resource pool, in response to the similarity between the first content tag and the content tag corresponding to the i-th candidate multimedia content reaching a first preset threshold, the i-th candidate multimedia content is determined as the second multimedia content; i is a positive integer.

5. The method according to claim 4, characterized in that The method further comprises: When performing a matching analysis on the jth candidate multimedia content in the candidate content resource pool, in response to the similarity between the first content tag and the content tag corresponding to the jth candidate multimedia content not reaching the first preset threshold, and the similarity between the content tag corresponding to the i-th candidate multimedia content and the content tag corresponding to the j-th candidate multimedia content reaches a second preset threshold, the j-th candidate multimedia content is determined as the second multimedia content; j is a positive integer, and i and j are different.

6. The method according to claim 2, characterized in that The acquiring the second multimedia content based on the preference tag includes: Acquire a candidate content resource pool, wherein the candidate content resource pool includes a plurality of candidate multimedia contents marked with a production list, wherein the production list includes the names of producers who participated in producing the candidate multimedia contents; In a case where the preference tag includes the first production list, in response to the overlap between the production list of the candidate multimedia content and the first production list meeting a preset list overlap requirement, at least one candidate multimedia content meeting the list overlap requirement is determined as the second multimedia content.

7. The method according to any one of claims 1 to 6, characterized in that: The playback information includes a plurality of sub-information, the kth sub-information includes the kth playback timestamp information and the column information to which the kth content belongs, and k is a positive integer; The obtaining the playback information of the first multimedia content includes: In the case of acquiring the kth sub-information, receiving the kth playback image and the kth playback channel information sent by the first client, wherein the kth playback image is an image captured when the first client plays the first multimedia content, and after the first client captures the kth playback image, identifies the kth playback image and determines the kth playback channel information; Determining k-th playback timestamp information of the first multimedia content based on the k-th playback image; The column information to which the kth content belongs is determined based on the kth playback channel information and the kth playback timestamp information, wherein the kth content column information is used to indicate the column to which the first multimedia content played by the first client in the kth sub-information belongs.

8. The method according to claim 7, characterized in that The determining the preference tag of the first account based on the playback information includes: In response to the number of the sub-information corresponding to the playback information reaching a preset number threshold, the multiple pieces of sub-information are counted to determine a target playback channel that meets the preset playback time requirement among the multiple pieces of sub-information; wherein the ratio of the playback time of the first multimedia content played on the first client through the target playback channel to the total playback time of the first multimedia content played on the first client meets the preset ratio requirement; Acquire a list of programs played by the target broadcast channel in a historical time period, wherein the list of programs includes at least one candidate program marked with a program label; Determine a column whose audience rating meets a preset audience rating requirement among the at least one candidate column as a target column; Determine that the preference tags of the first account include a column tag corresponding to the target column.

9. The method according to claim 8, characterized in that The acquiring the second multimedia content based on the preference tag includes: Acquire a candidate content resource pool, wherein the candidate content resource pool includes a plurality of candidate multimedia contents marked with content tags, wherein the content tags are used to indicate the subject of the candidate multimedia contents; In the case where the preference tag includes a column tag corresponding to the target column, in response to the existence of a content tag of candidate multimedia content matching the column tag, at least one candidate multimedia content matching the column tag is determined as the second multimedia content.

10. The method according to any one of claims 1 to 6, characterized in that: After obtaining the playback information of the first multimedia content in the first client, the method further includes: When the playback information indicates that the first multimedia content is paused during playback, determining a pause timestamp of the first multimedia content; Determining synchronous playback content of the first account based on the playback information, wherein the first multimedia content includes the synchronous playback content; Determining a start timestamp when playing the synchronously played content based on the pause timestamp, wherein the start timestamp refers to a position in a progress bar of the synchronously played content; The synchronously played content is recommended to the second client, wherein the second client user displays the synchronously played content on a recommendation page, and the second client plays the synchronously played content starting from the start and end timestamps.

11. The method according to any one of claims 1 to 6, characterized in that: The method further comprises: Acquire live broadcast information of the first client, where the live broadcast information is used to describe the third multimedia content played live by the first client; Determine a live broadcast preference tag of the first account based on the live broadcast information, where the live broadcast preference tag is used to indicate the preference of the first account for watching multimedia content in the live broadcast; Acquiring fourth multimedia content based on the live broadcast preference tag; At least one of the fourth multimedia content and the third multimedia content is recommended to the second client.

12. A multimedia content recommendation device, characterized in that: The device comprises: an acquisition module, configured to acquire playback information of a first multimedia content in a first client, wherein the playback information is used to indicate a column to which the first multimedia content played in the first client belongs, and a first account is logged in the first client; a preference determination module, configured to determine a preference tag of the first account based on the playback information, wherein the preference tag is a preference of the first account for playing multimedia content determined based on the playback information; a content acquisition module, configured to acquire second multimedia content based on the preference tag, wherein the second multimedia content is of a different content type from the first multimedia content; The content recommendation module is used to recommend the second multimedia content to the second client, wherein the second client is used to display the second multimedia content on a recommendation page, and an association relationship between the first client and the second client is established by logging into the first account.

13. A computer device, characterized in that: The computer device includes a processor and a memory, wherein the memory stores at least one program, and the at least one program is loaded and executed by the processor to implement the multimedia content recommendation method according to any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that: The storage medium stores at least one program, and the at least one program is loaded and executed by the processor to implement the multimedia content recommendation method according to any one of claims 1 to 11.

15. A computer program product, characterized in that The method comprises a computer program, which, when executed by a processor, implements the multimedia content recommendation method according to any one of claims 1 to 11.

Images