Family film and television recommendation method, device and equipment

Abstract

The invention relates to a family film and television recommendation method, device and equipment, and belongs to the technical field of video recommendation, and the method comprises the steps: obtaining local film data and user behavior data, obtaining film and television recommendation data through an artificial intelligence model based on the local film data and the user behavior data, and sending the film and television recommendation data to an operation server, the operation personnel can recommend the film and television to the user according to the film and television recommendation data; after the artificial intelligence model is accessed, the timely film recommendation information can be obtained, and then the local film data and the user behavior data are combined, so that the video content can be accurately recommended for the user, and the click rate is effectively increased.

Description

Technical Field

[0001] The present application relates to the technical field of video recommendation, and in particular to a method, device and equipment for recommending home movies and TV shows. Background Art

[0002] The recommended content of movies and TV shows in family movie-watching services generally comes from traditional operation modes or movie-watching algorithm operation modes. The traditional operation mode is to regularly arrange screens through the operation of fixed desktops, and send the arranged program desktops to the APP through the operation system for display of recommended content; the movie-watching algorithm operation mode is to recommend similar types of movies based on the user's current movie-watching records. The traditional operation mode is lagging, and it often starts operation and promotion only after it becomes popular, which will lose a large number of subscribers; the movie-watching algorithm operation mode recommends data based on existing movie-watching types. The data in the existing movie-watching types is outdated, and users still have messy clicks, which leads to deviations between the recommended movie-watching content and the content that users are interested in, resulting in inaccurate recommended content. Summary of the invention

[0003] To this end, the present invention provides a method, device and equipment for recommending home movies and TV shows to solve the problems in the prior art that the recommended content has a lag and the recommended viewing content deviates from the content that the user is interested in.

[0004] In a first aspect, a method for recommending home movies and TV shows is provided, the method comprising:

[0005] Acquire local basic data; the local basic data includes film data and user behavior data; the film data includes film data of all film and television works in the local database;

[0006] Obtaining film and television recommendation data through an artificial intelligence model based on the local basic data;

[0007] The film and television recommendation data is sent to the operation server so that the operation personnel can make film and television recommendations to users based on the film and television recommendation data.

[0008] Furthermore, obtaining film and television recommendation data through an artificial intelligence model based on the local basic data includes:

[0009] Based on the film data, film and television recommendation data is obtained through an artificial intelligence model.

[0010] or,

[0011] The film and television recommendation data is obtained based on the film data and the user behavior data through an artificial intelligence model.

[0012] Furthermore, the film and television data at least includes the film and television type and the film and television name, and the film and television recommendation data is obtained through the artificial intelligence model based on the film data, including:

[0013] Obtaining a first preset number of question statements according to a preset recommendation tag; the preset recommendation tag is a tag generated according to a preset recommendation vocabulary and a preset film and television type;

[0014] Inputting a first preset number of the question sentences into the artificial intelligence model to obtain a first preset number of answer content data;

[0015] The film and television title with the highest repetition rate in the reply content data is used as the pending film and television title;

[0016] If a film name identical to the pending film name is found in the film data, and the film type corresponding to the film name is identical to the preset film type, the pending film name is used as the target film name;

[0017] The film and television data corresponding to the target film and television name is used as the film and television recommendation data.

[0018] Furthermore, the method further comprises:

[0019] The reply content data including the target film or TV title is stored in the knowledge base of the artificial intelligence model.

[0020] Furthermore, obtaining film and television recommendation data through an artificial intelligence model based on the film data and the user behavior data includes:

[0021] Obtaining a second preset number of question statements according to the user behavior data;

[0022] Inputting the question sentence into the artificial intelligence model to obtain answer content data;

[0023] The reply content data is verified based on the film data, and if the reply content data passes the verification, the film and television recommendation data is obtained based on the reply content data.

[0024] Further, the reply content data is verified based on the film data, and if the reply content data passes the verification, the film and television recommendation data is obtained based on the reply content data, including:

[0025] The film and television titles with the highest repetition rate in the reply content data are used as the film and television titles to be verified;

[0026] The name of the film or TV program to be verified is used as a search condition to search in the film data. If a film or TV program that is the same as the name of the film or TV program to be verified is found, the reply content data is verified, and the name of the film or TV program to be verified is used as the approved film or TV program name.

[0027] The film and television data corresponding to the approved film and television name is used as the film and television recommendation data.

[0028] Furthermore, it also includes:

[0029] The film data is queried using the name of the film to be verified as a search condition. If no film or TV name identical to the film or TV name to be verified is found, the third-party shared resource is queried using the film or TV name to be verified as a search condition. If a film or TV name identical to the film or TV name to be verified is found, the reply content data is verified, and the film or TV name to be verified is used as the approved film or TV name;

[0030] The film and television data corresponding to the approved film and television name is obtained from the third-party shared resource as the film and television recommendation data, and the film data corresponding to the approved film and television name is stored in a local database.

[0031] Furthermore, obtaining film and television recommendation data through an artificial intelligence model based on the film data and the user behavior data includes:

[0032] Inputting the user behavior data into an artificial intelligence model to obtain a plurality of user tags;

[0033] Obtain third-party recommendation ratings and third-party recommendation data for film and television works in third-party shared resources;

[0034] The film and television recommendation data is obtained according to the plurality of user tags, the third-party recommendation scores and the third-party recommendation data.

[0035] In a second aspect, a home video recommendation device is provided, the device comprising:

[0036] An acquisition module, used to acquire local basic data; the local basic data includes film data and user behavior data; the film data includes film data of all film and television works in the local database;

[0037] A recommendation data generation module, used to obtain film and television recommendation data through an artificial intelligence model based on the local basic data;

[0038] The transfer module is used to send the film and television recommendation data to the operation server so that the operation personnel can make film and television recommendations to users based on the film and television recommendation data.

[0039] In a third aspect, a computer device is provided, the computer device comprising a processor and a memory, the memory storing a computer program, the computer program being loaded and executed by the processor to implement the above-mentioned method for recommending home movies and TV shows.

[0040] The present invention adopts the above technical solution and has at least the following beneficial effects:

[0041] Provided is a method, device and equipment for recommending home movies and TV shows. First, local movie data and user behavior data are obtained. Based on the local movie data and user behavior data, movie and TV show recommendation data is obtained through an artificial intelligence model. The movie and TV show recommendation data is sent to an operation server so that an operation staff can make movie and TV show recommendations to users based on the movie and TV show recommendation data. After accessing the artificial intelligence model, timely movie recommendation information can be obtained. Combined with the local movie data and user behavior data, accurate video content can be recommended to users, which can effectively increase the click-through rate.

[0042] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 is a flow chart of a method for recommending home movies and TV shows according to an exemplary embodiment of the present invention;

[0045] Figure 2 It is a flowchart of the existing recommendation algorithm;

[0046] Figure 3 is a flow chart of a method for recommending home movies and TV programs according to another exemplary embodiment of the present invention;

[0047] Figure 4 is a flow chart showing a method of verifying reply content data according to an exemplary embodiment of the present invention;

[0048] Figure 5 It is a schematic block diagram of a home video recommendation device shown in an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0050] Movie viewing services have always been a difficult problem raised by operations. Traditional operation desktops cannot meet the needs of operations. How to assist in improving movie viewing indicators and increase profitability has become the main problem that needs to be optimized. For movie viewing services, the existing technical solutions are: 1. Desktop touch point operation, based on the current popular movies and TV shows on the market, manual operation data is recommended on the first screen of the homepage; 2. Movie viewing algorithm operation, based on the user's current movie viewing records, similar types of movie recommendations are extended to medium and long-term predictions; However, the existing technology has many defects, for example, 1. The operation of desktop recommendation positions is currently lagging, and operations and promotions are often started only after the popularity has become popular, which will lose a large number of subscribers; 2. The recommended data based on the movie viewing type is messy, the click-through rate is not high, and the recommended data is outdated, and it cannot achieve the expected goals.

[0051] The movie viewing algorithm operation in the existing technology, such as Figure 2 As shown, Figure 2 This is a flowchart of the existing recommendation algorithm. First, when a user clicks on a movie through the recommendation position to watch a movie, the tracking module will be started on the APP side, and the movie data and user data clicked to watch will be uploaded to the local server as local basic data. Secondly, the local server will perform statistics based on the reported local basic data, summarize the data based on the movie data, make first-level recommendations based on the type, and make second-level recommendations based on the popularity of the movie ratings, and send the first-level and second-level recommendations to the operation server; at the same time, the local server generates corresponding historical records and collection records and other user behaviors based on the current data report, and binds it with the user data to send the user behavior data for movie viewing services to the operation server. Finally, based on the first-level and second-level recommendations and user behavior data as the overall movie viewing companion data, the user's face data is constructed to form the corresponding basic recommendation content. The operation staff is responsible for pushing the recommended content to the user's APP side. For example, recommendations can be made based on the user's viewing behavior, such as the most favorite categories, the content with the longest viewing time, etc.

[0052] The embodiment of the present application provides a method, device and equipment for recommending home movies and TV. First, local movie data and user behavior data are obtained. Based on the local movie data and user behavior data, movie and TV recommendation data is obtained through an artificial intelligence model, and the movie and TV recommendation data is sent to the operation server so that the operation personnel can recommend movies and TV to users based on the movie and TV recommendation data; the AI ​​model (artificial intelligence model) is combined for active service, which increases the exposure speed of current popular movie and TV content by 8-12 hours; the AI ​​model is used to count user behavior and analyze user topology, which is faster and more accurate. The traditional R&D code is used to count data. The time period for analyzing a user's behavior is usually three to six months. Through the big data AI model, the same effect can be achieved in only one month, which greatly improves efficiency. At the same time, the AI ​​model is connected and currently used as a search engine, because the current AI model on the market does not have a professional movie viewing model. If it is used directly, it will often return wrong information, answer the wrong question, and cannot be used commercially. However, we can use our own business as a basis to train our own AI model for movie viewing business, so that our own AI model has the underlying content foundation of film and television variety shows, and then later we can autonomously provide services for operations through AI capabilities.

[0053] The method in this application is described below through specific examples.

[0054] See also Figure 1 , Figure 1 is a flowchart of a method for recommending home video according to an exemplary embodiment of the present invention. Figure 1 , the method comprising:

[0055] Step S11, obtaining local basic data; the local basic data includes film data and user behavior data; the film data includes film data of all film and television works in the local database;

[0056] Step S12: Obtaining film and television recommendation data through an artificial intelligence model based on local basic data;

[0057] Step S13: Send the film and television recommendation data to the operation server so that the operation personnel can make film and television recommendations to users based on the film and television recommendation data.

[0058] It should be noted that the technical solution provided in this embodiment can be loaded into an existing film and television system or application in the form of a small program or a plug-in in practice, or in the form of a separate application, and the function of film and television data recommendation can be realized through an external interface. Applicable scenarios include but are not limited to: TV series video recommendation, movie video recommendation, variety show video recommendation and other scenarios that require recommended video content.

[0059] Specifically, behavioral data includes user behavior data such as historical click records, collection records, and film playback time; film and television recommendation data includes film and television introduction content such as film and television names, film and television cover images, film and television introductions, and film and television wonderful screenshots.

[0060] It can be understood that the home movie and TV recommendation method provided in this embodiment first obtains local movie data and user behavior data, and based on the local movie data and user behavior data, obtains movie and TV recommendation data through an artificial intelligence model, and sends the movie and TV recommendation data to the operation server so that the operation personnel can make movie and TV recommendations to users based on the movie and TV recommendation data; after accessing the artificial intelligence model, timely movie recommendation information can be obtained, and combined with the local movie data and user behavior data, it can accurately recommend video content to users, which can effectively increase the click-through rate.

[0061] In specific practice, "obtaining local basic data" in step S11 includes: obtaining film data and user behavior data from the local server, the film data includes the film and television data of all film and television works in the licensee's media library, and the user behavior data includes user behaviors such as click records, collection records and film playback time.

[0062] See also Figure 3 , Figure 3 It is a flowchart of a method for recommending home movies and TV shows shown in another exemplary embodiment of the present invention. In specific practice, step S12 "obtaining movie and TV show recommendation data through an artificial intelligence model based on local basic data" includes: obtaining movie and TV show recommendation data through an artificial intelligence model based on movie data, or obtaining movie and TV show recommendation data through an artificial intelligence model based on movie data and user behavior data.

[0063] It should be noted that the artificial intelligence model is an existing artificial intelligence model, such as Doubao, Wenxinyiyan, etc., which are artificial intelligence software that are good at question-and-answering.

[0064] In specific practice, the film and television data at least includes the film and television type and the film and television name, and the film and television recommendation data is obtained based on the film data through an artificial intelligence model, including: obtaining a first preset number of question statements based on a preset recommendation label; the preset recommendation label is a label generated based on a preset recommendation vocabulary and a preset film and television type; inputting the first preset number of question statements into the artificial intelligence model to obtain a first preset number of reply content data; taking the film and television name with the highest repetition rate in the reply content data as the pending film and television name; if the film and television name identical to the pending film and television name is queried in the film data, and the film and television type corresponding to the film and television name is the same as the preset film and television type, then taking the pending film and television name as the target film and television name; taking the film and television data corresponding to the target film and television name as the film and television recommendation data.

[0065] It should be noted that the first preset number is set according to the recommendation accuracy requirement; the preset recommendation words can be the most popular, most watched, most searched, and other recommendation words corresponding to the video recommendation section in the video application, and the preset film and television types can be existing film and television types such as movies, TV series, and variety shows, and can be specific to film and television tags, for example, fantasy movies.

[0066] Specifically, by making a scheduled task script, for example, starting the script at 8:30 pm, 10:30 pm, 12:00 pm, and 4:00 am every day, searching for the most popular film and television variety shows of the day, taking the most popular as the preset recommended vocabulary, and taking variety show as the film and television type, generating a limited number of question sentences and inputting them into the artificial intelligence model, obtaining a limited number of reply content data, and taking the film and television names with the highest repetition rate in the reply content as the pending film and television names, and finding the film and television names that are the same as the pending film and television names in the film data, and the film and television types corresponding to the film and television names are the same as the preset film and television types, then taking the pending film and television names as the target film and television names, and taking the film and television data corresponding to the target film and television names as the film and television recommendation data, and recommending the film and television recommendation data to the user APP at 8:30 pm, 10:30 pm, 12:00 pm, and 4:00 am every day.

[0067] It can be understood that the home video recommendation method provided in this embodiment can realize active video recommendation based on real-time video, effectively avoid outdated recommended content, make the recommended content consistent with viewing hot spots, and effectively increase the user's click rate.

[0068] In specific practice, the reply content data containing the target film and television names is stored in the knowledge base of the artificial intelligence model, so that the artificial intelligence model can be effectively trained so that the next reply will be more accurate.

[0069] In specific practice, film and television recommendation data is obtained through an artificial intelligence model based on film data and user behavior data, including: obtaining a second preset number of question statements based on user behavior data; inputting the question statements into the artificial intelligence model to obtain answer content data; verifying the answer content data based on the film data, and if the answer content data passes the verification, obtaining film and television recommendation data based on the answer content data.

[0070] It should be noted that a second preset number of question statements are obtained based on user behavior data, and the second preset data is set according to the required accuracy; question statements can be generated based on user historical click records. For example, if the most clicked film and television type in the user's click records is comedy movies, then the generated question statements are: recommend a newly released comedy movie, recommend the most popular comedy movie, etc.

[0071] It is understandable that the home video and movie recommendation method provided in this embodiment can make video and movie recommendations based on user behavior habits, making the recommended content more accurate.

[0072] See also Figure 4 , Figure 4 is a flowchart showing a method of verifying reply content data according to an exemplary embodiment of the present invention, see Figure 4 The method includes: verifying the reply content data based on the film data, including: using the film and television name with the highest repetition rate in the reply content data as the film and television name to be verified; using the film and television name to be verified as a search condition to query in the film data, if the film and television name that is the same as the film and television name to be verified can be queried, then the reply content data is verified, and the film and television name to be verified is used as the approved film and television name; using the film and television data corresponding to the approved film and television name as the film and television recommendation data.

[0073] Specifically, it also includes: using the name of the film or television to be verified as a search condition to query in the film data, if the film or television name that is identical to the name of the film or television to be verified is not found, then using the film or television name to be verified as a search condition to query in the third-party shared resources, if the film or television name that is identical to the name of the film or television to be verified is found, then replying that the content data verification is passed, and using the film or television name to be verified as the approved film or television name; obtaining the film and television data corresponding to the approved film or television name from the third-party shared resources as film and television recommendation data, and storing the film data corresponding to the approved film and television name in the local database.

[0074] It can be understood that the home movie and TV recommendation method provided in this embodiment, by further verifying the recommended content, effectively reduces the situation where the wrong content is recommended due to the artificial intelligence model's non-standard response to the movie and TV name.

[0075] In specific practice, it also includes: inputting user behavior data into an artificial intelligence model to obtain multiple user tags; obtaining third-party recommendation scores and third-party recommendation data of film and television works in third-party shared resources; and obtaining film and television recommendation data based on multiple user tags, third-party recommendation scores and third-party recommendation data.

[0076] It should be noted that user tags can be: like science fiction movies, like Chinese movies..., and the third-party shared library will search for an item that matches the user tag in the recommended film and television data from the highest to the lowest ratings. The third-party data corresponding to the corresponding film and television title will be used as the target video recommendation data and sent to the operation server for selective push to customers later.

[0077] It can be understood that the home film and television recommendation method provided in this embodiment generates recommendation data through artificial intelligence models and third-party shared resources, effectively reducing the dependence of the recommendation data service on the local server.

[0078] See also Figure 5 , Figure 5is a schematic block diagram of a home video recommendation device according to an exemplary embodiment of the present invention. Figure 5 , a recommended device 100 for home video, comprising:

[0079] The acquisition module 101 is used to acquire local basic data; the local basic data includes film data and user behavior data; the film data includes film data of all film and television works in the local database;

[0080] The recommendation data generation module 102 is used to obtain film and television recommendation data through an artificial intelligence model based on local basic data;

[0081] The transfer module 103 is used to send the film and television recommendation data to the operation server so that the operation staff can make film and television recommendations to users based on the film and television recommendation data.

[0082] It should be noted that the home video recommendation device provided in this embodiment is applicable to scenarios including but not limited to: TV series video recommendation, movie video recommendation, variety show video recommendation and other scenarios that require recommended video content.

[0083] It can be understood that the home movie and TV recommendation device provided in this embodiment first obtains local movie data and user behavior data, and based on the local movie data and user behavior data, obtains movie and TV recommendation data through an artificial intelligence model, and sends the movie and TV recommendation data to the operation server so that the operation personnel can make movie and TV recommendations to users based on the movie and TV recommendation data; after accessing the artificial intelligence model, timely movie recommendation information can be obtained, and combined with the local movie data and user behavior data, it can accurately recommend video content to users, which can effectively increase the click-through rate.

[0084] In a specific embodiment, a computer device is provided. The computer device includes a processor and a memory. The memory stores a computer program. The computer program is loaded and executed by the processor to implement the above-mentioned method for recommending home movies and TV shows.

[0085] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

[0086] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0087] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0088] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for recommending home movies and TV shows, characterized in that: The method comprises: Acquire local basic data; the local basic data includes film data and user behavior data; the film data includes film data of all film and television works in the local database; Obtaining film and television recommendation data through an artificial intelligence model based on the local basic data; The film and television recommendation data is sent to the operation server so that the operation personnel can make film and television recommendations to users based on the film and television recommendation data.

2. The method according to claim 1, characterized in that: The obtaining of film and television recommendation data through an artificial intelligence model based on the local basic data includes: Based on the film data, film and television recommendation data is obtained through an artificial intelligence model. or, The film and television recommendation data is obtained based on the film data and the user behavior data through an artificial intelligence model.

3. The method according to claim 2, characterized in that The film and television data at least includes the film and television type and the film and television name, and the film and television recommendation data is obtained by using the artificial intelligence model based on the film data, including: Obtaining a first preset number of question statements according to a preset recommendation tag; the preset recommendation tag is a tag generated according to a preset recommendation vocabulary and a preset film and television type; Inputting a first preset number of the question sentences into the artificial intelligence model to obtain a first preset number of answer content data; The film and television name with the highest repetition rate in the reply content data is used as the pending film and television name; If a film name identical to the pending film name is found in the film data, and the film type corresponding to the film name is identical to the preset film type, the pending film name is used as the target film name; The film and television data corresponding to the target film and television name is used as the film and television recommendation data.

4. The method according to claim 3, characterized in that The method further comprises: The reply content data including the target film or TV title is stored in the knowledge base of the artificial intelligence model.

5. The method according to claim 2, characterized in that: The obtaining of film and television recommendation data through an artificial intelligence model based on the film data and the user behavior data includes: Obtaining a preset number of question statements based on the user behavior data; Inputting the question sentence into the artificial intelligence model to obtain answer content data; The reply content data is verified based on the film data, and if the reply content data passes the verification, the film and television recommendation data is obtained based on the reply content data.

6. The method according to claim 5, characterized in that The step of verifying the reply content data based on the film data, and obtaining the film and television recommendation data based on the reply content data if the reply content data passes the verification, includes: The film and television titles with the highest repetition rate in the reply content data are used as the film and television titles to be verified; The name of the film or TV program to be verified is used as a search condition to search in the film data. If a film or TV program that is the same as the name of the film or TV program to be verified is found, the reply content data is verified, and the name of the film or TV program to be verified is used as the approved film or TV program name. The film and television data corresponding to the approved film and television name is used as the film and television recommendation data.

7. The method according to claim 6, characterized in that Also includes: The film data is queried using the name of the film to be verified as a search condition. If no film or TV name identical to the film or TV name to be verified is found, the third-party shared resource is queried using the film or TV name to be verified as a search condition. If a film or TV name identical to the film or TV name to be verified is found, the reply content data is verified, and the film or TV name to be verified is used as the approved film or TV name; The film and television data corresponding to the approved film and television name is obtained from the third-party shared resource as the film and television recommendation data, and the film data corresponding to the approved film and television name is stored in a local database.

8. The method according to claim 2, characterized in that: Also includes: Inputting the user behavior data into an artificial intelligence model to obtain a plurality of user tags; Obtain third-party recommendation ratings and third-party recommendation data for film and television works in third-party shared resources; The film and television recommendation data is obtained according to the plurality of user tags, the third-party recommendation scores and the third-party recommendation data.

9. A home video recommendation device, characterized in that: The device comprises: An acquisition module, used to acquire local basic data; the local basic data includes film data and user behavior data; the film data includes film data of all film and television works in the local database; A recommendation data generation module, used to obtain film and television recommendation data through an artificial intelligence model based on the local basic data; The transfer module is used to send the film and television recommendation data to the operation server so that the operation personnel can make film and television recommendations to users based on the film and television recommendation data.

10. A computer device, characterized in that: The computer device includes a processor and a memory, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the home video and movie recommendation method according to any one of claims 1 to 8.

Images