Content recommendation method and related device

Abstract

The invention provides a content recommendation method and a related device. The method comprises the steps that a big disk object data set of a media platform is acquired, the media platform comprises a plurality of media contents, the big disk object data set comprises object data of a plurality of big disk objects, and each piece of object data comprises object features of multiple dimensions; determining a screening condition of each dimension based on the advertisement type of the target advertisement; screening out a plurality of target objects meeting screening conditions from the large-disk object data set; and according to the media playing conditions of the big disk object and the target object in the media platform, determining the adaptation degree of the plurality of media contents and the target advertisement. The scheme can be used for assisting an advertiser in advertisement putting selection.

Description

Technical Field

[0001] This application relates to the technical field of advertisement recommendation, and particularly to a content recommendation method and related devices. Background Art

[0002] Media playback platforms include various types of players and software for playing various media contents. Advertisers can use the media platforms to display advertisement information to target audiences. Advertising contents can be inserted into various media contents. For advertisers, they usually consider whether to place advertisements under a media content based on the number of times the media content is played (Video View, VV). However, this method will result in most advertisements being placed on top contents. For advertisers, the advertising cost is relatively high. For media playback platforms, it is difficult to fully sell non-top content resources. Therefore, how to provide a content recommendation method that comprehensively considers advertising contents and media contents to determine the value of media contents for an advertising content and assist advertisers in making advertising placement choices is a technical problem that needs to be solved urgently by those skilled in the art. Summary of the Invention

[0003] Embodiments of this application provide a content recommendation method and related devices for assisting advertisers in making advertising placement choices.

[0004] A first aspect of this application provides a content recommendation method, including:

[0005] Obtain a large-scale object dataset of a media platform. The media platform includes multiple media contents. The large-scale object dataset includes object data of multiple large-scale objects, and each object data includes object features in multiple dimensions;

[0006] Determine the screening conditions for each dimension based on the advertisement type of a target advertisement;

[0007] Screen out multiple target objects that meet the screening conditions from the large-scale object dataset;

[0008] Determine the adaptation degree of multiple media contents to the target advertisement respectively according to the media playback situations of the large-scale objects and the target objects in the media platform.

[0009] In a possible implementation method, after screening out multiple target objects that meet the screening conditions from the large-scale object dataset, it further includes:

[0010] Establish a target object dataset based on the multiple target objects. The target object dataset includes object data of multiple target objects;

[0011] Establish a target object chart according to the target object dataset. The target object chart indicates the distribution of target objects in multiple dimensions.

[0012] In a possible implementation method, after establishing a target object chart based on the target object dataset, it further includes:

[0013] Based on the overall market object dataset and the target object dataset, determine the target group index in multiple dimensions. The target group index indicates the ratio of the first distribution to the second distribution. The first distribution indicates the distribution of the target objects in multiple dimensions, and the second distribution indicates the distribution of the overall market objects in multiple dimensions;

[0014] Display the target group index in the target object chart.

[0015] In a possible implementation method, determine the screening conditions for each dimension based on the advertisement type of the target advertisement, including:

[0016] Based on the advertisement type of the target advertisement, determine the corresponding consumer objects;

[0017] Based on the consumer objects, determine the screening conditions.

[0018] In a possible implementation method, determine the adaptation degree of multiple media contents to the target advertisement according to the media playback situations of the overall market objects and the target objects in the media platform respectively, including:

[0019] Determine the playback metrics of each media content according to the media playback situations of the overall market objects and the target objects in the media platform respectively;

[0020] Rank the media contents based on each playback metric to obtain the ranking results of each media content under each playback metric;

[0021] Obtain the weights of each playback metric;

[0022] Combine the weights with the ranking results to obtain the value scores of each media content. The value scores are used to indicate the adaptation degree of the media contents to the target advertisement.

[0023] In a possible implementation method, determine the adaptation degree of multiple media contents to the target advertisement according to the media playback situations of the overall market objects and the target objects in the media platform respectively, including:

[0024] According to the playback situations, output the adaptation degrees of multiple media categories to the target advertisement. Each media category includes multiple media themes, and each media theme includes multiple media contents;

[0025] In response to a selection operation on a specified category, output the adaptation degrees of multiple media themes under the specified category to the target advertisement. The specified category belongs to the multiple media categories;

[0026] In response to a selection operation for a specified theme, determine the degree of fit between each media content under the specified theme and the target advertisement, where the specified theme belongs to multiple media themes.

[0027] In a possible implementation method, after obtaining the overall market object dataset of the media platform, it further includes:

[0028] Screen out the viewer objects that have accessed the specified content from the overall market object dataset, where the specified content belongs to multiple media contents;

[0029] Based on the click-through rates of the viewer objects for multiple advertisement contents, determine the degree of interest of the viewer objects in each advertisement content.

[0030] The second aspect of this application provides a content recommendation device, including:

[0031] An acquisition module, configured to acquire the overall market object dataset of the media platform, where the media platform includes multiple media contents, and the overall market object dataset includes the object data of multiple overall market objects, and each object data includes object features of multiple dimensions;

[0032] A condition determination module, configured to determine the screening conditions for each dimension based on the advertisement type of the target advertisement;

[0033] A screening module, configured to screen out multiple target objects that meet the screening conditions from the overall market object dataset;

[0034] An adaptation module, configured to determine the degree of fit between multiple media contents and the target advertisement respectively according to the media playback situations of the overall market objects and the target objects in the media platform.

[0035] In a possible implementation method, it further includes:

[0036] A chart building module, configured to establish an overall market object dataset based on multiple target objects, where the overall market object dataset includes the object data of multiple target objects; establish an overall market object chart based on the overall market object dataset, and the overall market object chart indicates the distribution of the target objects in multiple dimensions.

[0037] In a possible implementation method, it further includes:

[0038] The chart building module is further configured to determine the target group index in multiple dimensions based on the overall market object dataset and the overall market object dataset, where the target group index indicates the ratio of the first distribution to the second distribution, the first distribution indicates the distribution of the target objects in multiple dimensions, and the second distribution indicates the distribution of the overall market objects in multiple dimensions; display the target group index in the overall market object chart.

[0039] In a possible implementation method, a condition determination module is specifically configured to determine a corresponding consumption object based on the advertisement type of the target advertisement; and determine a screening condition based on the consumption object.

[0040] In a possible implementation method, an adaptation module is specifically configured to determine the playback metrics of each media content according to the media playback situations of the overall market object and the target object in the media platform respectively; sort the media content based on each playback metric to obtain the ranking result of each media content under each playback metric; obtain the weight of each playback metric; and combine the weight with the ranking result to obtain the value score of each media content, where the value score is used to indicate the adaptation degree of the media content to the target advertisement.

[0041] In a possible implementation method, an adaptation module is specifically configured to output the adaptation degrees of multiple media categories to the target advertisement according to the playback situation, where each media category includes multiple media themes, and each media theme includes multiple media contents; in response to a selection operation on a specified category, output the adaptation degrees of multiple media themes under the specified category to the target advertisement, where the specified category belongs to the multiple media categories; and in response to a selection operation on a specified theme, determine the adaptation degree of each media content under the specified theme to the target advertisement, where the specified theme belongs to the multiple media themes.

[0042] In a possible implementation method, it further includes:

[0043] A media content analysis module filters out the audience objects that have accessed the specified content from the overall market object dataset, where the specified content belongs to the multiple media contents; and determines the degree of interest of the audience objects in each advertisement content based on the click-through rates of the audience objects on the multiple advertisement contents.

[0044] A third aspect of the present application provides a computer device, including: a memory and a processor;

[0045] The memory stores instructions, and when the instructions run on the processor, the methods of the above aspects are executed.

[0046] A fourth aspect of the present application provides a computer-readable storage medium, in which instructions are stored, and when the instructions run on a computer device, the computer device is caused to execute the methods of the above aspects.

[0047] A fifth aspect of the present application provides a computer program product or a computer program, where the computer program product or the computer program includes instructions, and the instructions are stored in a computer-readable storage medium. The processor of the computer device reads the instructions from the computer-readable storage medium, and the processor executes the instructions, causing the computer device to execute the methods provided in the above aspects.

[0048] As can be seen from the above technical solutions, the embodiments of the present application have the following advantages:

[0049] The present application provides a content recommendation method and related devices. By obtaining the object features of the overall market objects in various dimensions on the media platform, then screening the overall market objects based on the advertisement type of the target advertisement to obtain the target objects, and then determining the adaptation degree of the media content to the advertisement content based on the media playback situations of the target objects and the overall market objects, advertisers can use this adaptation degree as a reference for advertisement placement. The content recommendation method provided in this embodiment comprehensively considers the advertisement content and the media content, rather than only considering the playback times of the media content, which is beneficial to realizing more efficient sale of media content resources and improving the conversion rate of advertisement content. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is an application environment diagram of the content recommendation method in the embodiment of the present application;

[0051] Figure 2 It is a flowchart of the content recommendation method provided in the embodiment of the present application;

[0052] Figure 3 It is a flowchart of the content recommendation method provided in the embodiment of the present application;

[0053] Figures 4a to 4f It is a schematic interface diagram of Tool 1 provided in the embodiment of the present application;

[0054] Figures 5a to 5c It is a schematic interface diagram of Tool 2 provided in the embodiment of the present application;

[0055] Figure 6a and Figure 6b It is a schematic interface diagram of Tool 3 provided in the embodiment of the present application;

[0056] Figure 7a and Figure 7b It is a schematic content recommendation diagram of Scenario 1 provided in the embodiment of the present application;

[0057] Figure 8 It is a schematic content recommendation diagram of Scenario 2 provided in the embodiment of the present application;

[0058] Figure 9 It is a schematic content recommendation diagram of Scenario 3 provided in the embodiment of the present application;

[0059] Figure 10 It is a schematic content recommendation diagram of Scenario 4 provided in the embodiment of the present application;

[0060] Figure 11 It is a schematic content recommendation diagram of Scenario 5 provided in the embodiment of the present application;

[0061] Figure 12 A schematic diagram of an embodiment of the content recommendation device in the embodiments of the present application;

[0062] Figure 13 A schematic diagram of the server structure provided by the embodiments of the present application. Specific implementation manners

[0063] The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present application are used to distinguish similar objects and do not necessarily need to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "correspond to" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily need to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0064] A media playback platform is a tool for playing various types of media content, including various types of players and software. These platforms usually support multiple media formats, such as audio, video, animation, etc., enabling users to conveniently play and watch various media content. For advertisers, the media playback platform is an important promotional channel. They can display advertising information on the media playback platform to deliver advertisements to the target audience. By selecting the appropriate media playback platform and setting the appropriate placement strategy, advertisers can attract more potential customers and improve brand awareness and sales performance.

[0065] Media intellectual property (IP) refers to self-media with intellectual property rights. Such self-media can obtain traffic tilt, high popularity, produce valuable content, and brand influence from the platform. For advertisers, the effect and return rate of advertisements are relatively important in the placement strategy. Therefore, they usually choose popular media IPs for placement. Popular IPs usually refer to media IPs with a relatively high VV. However, if advertisers only consider the VV level, it will lead to concentrated placement on top content. For advertisers, the advertising placement cost is relatively high. For media playback platforms, non-top content resources are difficult to sell fully.

[0066] To solve the above problems, the embodiments of the present application provide a content recommendation method that comprehensively considers advertising content and media content to determine the degree of adaptation of media content to the advertising content, so as to achieve efficient and accurate recommendation of advertising content, which is conducive to realizing more efficient sale of media content resources and improving the conversion rate of advertising content.

[0067] To facilitate the understanding of the technical solutions provided by the embodiments of the present application, some key terms used in the embodiments of the present application are explained here:

[0068] Overall market object: The object refers to the user who watches media content. The overall market object refers to all objects without any restrictive conditions, specifically all registered objects under the media playback platform.

[0069] Target object: It refers to the object that the advertiser hopes the advertisement reaches.

[0070] Target Audience (TA) concentration: It refers to the proportion of the target object in the overall market object.

[0071] Penetration TA concentration: It refers to the proportion of the target object in the start-up UV (page views) of the overall market object.

[0072] Completion TA concentration: It refers to the proportion of the target object in the effective completion UV (completion volume) of the overall market object.

[0073] Completion rate: It refers to the proportion of the effective completion UV (completion volume) in the start-up UV (page views) of the object.

[0074] Completion rate TGI: Completion rate of the target object / Completion rate of the overall market.

[0075] Object characteristics: It refers to the characteristics of the object such as gender, age, consumption level, city, education level, life status, etc.

[0076] Media IP: It refers to the copyright corresponding to the media content. For example, for the TV drama "Love and Redemption", the media IP includes not only the TV drama itself, but also other video content related to "Love and Redemption" such as behind-the-scenes footage, related variety shows, and actor interviews.

[0077] Media category: It refers to the category to which the media content belongs. For example, when the media content is video content, the media category includes TV dramas, movies, variety shows, animations, etc.

[0078] Media theme: It refers to the theme corresponding to the media content. For example, for the TV drama "Love and Redemption", the main theme is love, and the secondary themes are ancient costumes and fantasy.

[0079] VV: It refers to the number of plays.

[0080] Viewing volume: Specifically, it refers to the number of unique visitors (UV) at the start of playback, which means the number of objects with more than 0 media content playback times.

[0081] Completion volume: Specifically, it refers to the effective completion UV, which means the number of objects with a media content playback completion rate not lower than a certain ratio. That is, the media content playback completion rate = total media content playback duration / total media content physical duration. When the media content is video content, the completion volume can be defined as the number of objects with a video content playback completion rate not lower than 50%.

[0082] The viewing volume specifically includes IP viewing volume, category viewing volume, and theme viewing volume. The completion volume specifically includes IP completion volume, category completion volume, and theme completion volume. Among them, the IP viewing volume refers to the number of objects with more than 0 IP content playback times; the IP completion volume refers to the number of objects with an IP content playback completion rate not lower than 50%; the category viewing volume and theme viewing volume refer to the number of objects with more than 0 playback times of at least one IP content under the category or theme; the category completion volume and theme completion volume refer to the number of objects with an IP content playback completion rate not lower than 50% under the category or theme.

[0083] Next, the content recommendation method and its related devices provided by the embodiments of the present application will be introduced in detail.

[0084] For ease of understanding, please refer to Figure 1 , Figure 1 which is the application environment diagram of the content recommendation method in the embodiments of the present application. As Figure 1 shown, the content recommendation method in the embodiments of the present application is applied to a content recommendation system. The content recommendation system includes: a server and a terminal device; among them, the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing 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 network (CDN), and big data and artificial intelligence platforms. The terminal device can be a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart TV, a smart watch, a vehicle-mounted terminal, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected through wired or wireless communication methods, and the embodiments of the present application do not make any restrictions here.

[0085] This method is applied to a business investment promotion recommendation scenario based on a media platform. The media platform includes multiple media contents. This method is used to recommend suitable media contents for a target advertisement for advertisement placement. Next, this method will be introduced from the perspective of the server:

[0086] The server first obtains the large-scale object dataset of the media platform. The large-scale object dataset includes the object data of multiple large-scale objects, and each object data includes object features in multiple dimensions. Each object feature is used to indicate the object category of the corresponding object in the current dimension;

[0087] Secondly, the server determines the screening conditions for each dimension based on the advertisement type of the target advertisement. For example, if the advertisement type of the target advertisement is a women's clothing advertisement, then the screening condition in the gender dimension is female;

[0088] Then, the target objects are screened out from the large-scale object dataset according to the screening conditions. As in the above example: when the advertisement type of the target advertisement is a women's clothing advertisement, the corresponding target objects are the female objects in the large-scale objects;

[0089] Finally, based on the media playback situations of the large-scale objects and the target objects in the media platform, the adaptation degree of multiple media contents in the media platform to the target advertisement is determined. This adaptation degree can be used by advertisers to screen suitable media contents for advertising placement.

[0090] It can be understood that the above method is executed by a computer device, and this computer device can be Figure 1 the terminal device or server shown in Figure 2 Figure 2 FIG.

[0091] 201. Obtain the large-scale object dataset of the media platform. The media platform includes multiple media contents. The large-scale object dataset includes the object data of multiple large-scale objects, and each object data includes object features in multiple dimensions.

[0092] It can be understood that the embodiments of the present application are used to recommend suitable media contents for the target advertisement for advertising placement. Therefore, it is first necessary to obtain the large-scale object dataset of the media platform, and the media platform includes multiple media contents.

[0093] The large-scale object dataset includes the object data of multiple large-scale objects. Each object data includes object features in multiple dimensions. The object feature is used to indicate the object classification in different dimensions. For example, the multiple dimensions include one or more of the following: gender, age, city level, life status, education level, consumption level, and work industry. It can be understood that the data source of the large-scale object dataset can be predicted based on the behaviors of registered objects in the media platform, or obtained through questionnaires for registered objects. The present application does not limit this data source.

[0094] ​Furthermore, based on the above object characteristics, the dimensions can also include nine major commercial user groups. The nine major commercial user groups refer to dividing user objects into nine categories through dimensions such as gender, age group, city level, consumption ability, and life status: Generation Z, refined moms, small-town youths, small-town middle-aged and elderly, urban silver-haired, new-generation white-collar workers, senior middle-class, urban blue-collar workers, and senior blue-collar workers.

[0095] For example, for object A, its corresponding object data can be represented in the form of a vector, that is, object A = {25 - 29, male, new first-tier city, unmarried, undergraduate, medium, finance, unknown, new-generation white-collar worker}, indicating that object A's age belongs to the range of 25 to 29 years old, gender is male, city level is a new first-tier city, life status is unmarried, education level is undergraduate, consumption level is medium, work industry is finance, and it belongs to the new-generation white-collar worker among the nine major commercial user groups.

[0096] 202. Determine the screening conditions for each dimension based on the ad type of the target ad.

[0097] It can be understood that determining the screening conditions for each dimension based on the ad type of the target ad means determining the target audience according to the ad type of the target ad, thereby determining the specified object characteristics under each dimension to more precisely place ads and improve the ad effect and return on investment.

[0098] Specifically, corresponding screening conditions can be set according to the characteristics of the target audience of the ad, such as age, gender, region, hobbies, etc., to ensure that the ad can accurately reach the target audience.

[0099] Furthermore, the screening conditions can also be determined according to the data source. For example, the object login device type (such as the placement platform, including Android platform, IOS platform, Windows platform, Mac platform, etc.), data collection time period, etc. For example, when the ad content is a mobile game, objects with a mobile platform login can be screened; when the ad content is related to the current hot topic, data with a data collection time period of the last month can be screened.

[0100] By determining the screening conditions for each dimension, ads can be placed more precisely, improving the ad coverage of target users and the overall ad click-through rate, thereby enhancing brand awareness and sales performance. At the same time, the ad strategy and placement plan can also be continuously adjusted according to the changes and optimizations of the screening conditions to adapt to market changes and changes in object needs.

[0101] 203. Screen out multiple target objects that meet the screening conditions from the overall market object dataset.

[0102] After determining the screening conditions, objects that meet the screening conditions can be screened out from the overall market object dataset, and this object is the target object of the target ad.

[0103] For example, if the target advertisement is targeted at female objects aged 20 - 35, then the screening conditions can be set as female objects aged between 20 - 35 and with the gender being female. By screening out the objects that meet these conditions from the large - scale object dataset, the target objects of the target advertisement can be found. The screened target objects are an important basis for advertising placement because these objects are more matched with the target audience of the advertisement and are more likely to be interested in and respond to the advertisement. Therefore, by determining the screening conditions and screening out the target objects, advertisements can be placed more precisely, improving the advertising effect and return on investment.

[0104] 204. Determine the degree of adaptation of multiple media contents to the target advertisement respectively according to the media playback situations of the large - scale objects and the target objects in the media platform.

[0105] After determining the large - scale objects and the target objects, the degree of adaptation of each media content to the target advertisement can be determined respectively according to the media playback situations of these objects in the media platform.

[0106] It can be understood that the media playback situation refers to various playback metrics of media content, such as the number of visits, the completion rate, the TA concentration, the penetration TA concentration, the completion TA concentration, the completion rate, the completion rate TGI, etc. The above - mentioned playback metrics can be statistically calculated according to the quantities of the large - scale objects and the target objects, so as to obtain the values of each media content under the above - mentioned playback metrics, and then sort the multiple media contents to determine the degree of adaptation of each media content to the target advertisement.

[0107] The content recommendation method provided by the embodiments of the present application obtains the object characteristics of the large - scale objects in the media platform in various dimensions, then screens the large - scale objects based on the advertisement type of the target advertisement to obtain the target objects, and then determines the degree of adaptation of the media content to the advertisement content based on the media playback situations of the target objects and the large - scale objects. Advertisers can use this degree of adaptation as a reference for advertising placement. The content recommendation method provided by this embodiment comprehensively considers the advertisement content and the media content, rather than only considering the number of plays of the media content, which is conducive to realizing more efficient sale of media content resources and improving the conversion rate of advertisement content.

[0108] In the Figure 2 corresponding optional embodiment of the content recommendation method provided by the embodiment of the present application, please refer to Figure 3 , Figure 3 which is the flowchart of the content recommendation method provided by the embodiment of the present application. This method is executed by a computer device, and the computer device can be Figure 1 the terminal device or server shown in

[0109] 301. Obtain the overall market object dataset of the media platform. The media platform includes multiple media contents. The overall market object dataset includes object data of multiple overall market objects, and each object data includes object features in multiple dimensions.

[0110] In this embodiment, step 301 is similar to step 201 in the corresponding Figure 2 embodiment above, and will not be elaborated here.

[0111] 302. Based on the advertisement type of the target advertisement, determine the corresponding consumer objects.

[0112] 303. Based on the consumer objects, determine the screening conditions.

[0113] In this embodiment, steps 302 and 303 correspond to step 202 in the corresponding Figure 2 embodiment above.

[0114] It can be understood that based on the advertisement type of the target advertisement, the corresponding consumer objects can be determined. For example, when the target advertisement is a certain furniture brand, based on the industry analysis report of the furniture industry, three main consumer groups in the furniture industry can be determined, namely: newly graduated college students, house-owning objects aged 25 - 45, and high-consumption objects aged 45 - 55. Then, according to the above 3 object groups, the screening conditions can be determined respectively:

[0115] Group 1: Newly graduated college student group, the screening conditions are: Age: between 22 and 25 years old; Education level: undergraduate or master.

[0116] Group 2: House-owning object group aged 25 - 45, the screening conditions are: Age: between 25 and 45 years old; Life status: married or having children.

[0117] Group 3: High-consumption object group aged 45 - 55, the screening conditions are: Age: between 45 and 55 years old; Consumption level: high.

[0118] 304. Screen out multiple target objects that meet the screening conditions from the overall market object dataset.

[0119] In this embodiment, step 304 is similar to step 203 in the corresponding Figure 2 embodiment above, and will not be elaborated here.

[0120] 305. Based on multiple target objects, establish a target object dataset. The target object dataset includes object data of multiple target objects.

[0121] 306. Establish a target object chart according to the target object dataset. The target object chart indicates the distribution of target objects in multiple dimensions.

[0122] In this embodiment, after obtaining multiple target objects, a target object dataset can be established based on the multiple target objects. The target object dataset includes object data of multiple target objects. It can be understood that each object data also includes object features in multiple dimensions.

[0123] After obtaining the object data of multiple target objects, according to the number of objects with different object features in each dimension, a target object chart can be constructed. The target object chart is used to indicate the distribution of target objects in each dimension. For example, after screening, the number of target objects is N. In the gender dimension, among the target objects, there are x males, y females, and z objects with unknown genders. It can be understood that N = x + y + z. Then an object chart in the gender dimension can be constructed, for example, in the form of a histogram, to reflect the gender distribution among the target objects.

[0124] By constructing the target object chart, the distribution of the group of objects interested in the advertisement content can be clearly and intuitively shown to the advertiser. This information can help the advertiser better understand the characteristics and needs of the target objects, so as to formulate more accurate advertising strategies and placement plans.

[0125] 307. Based on the overall market object dataset and the target object dataset, determine the target group index in multiple dimensions. The target group index indicates the ratio of the first distribution to the second distribution. The first distribution indicates the distribution of target objects in multiple dimensions, and the second distribution indicates the distribution of overall market objects in multiple dimensions.

[0126] 308. Display the target group index in the target object chart.

[0127] It can be understood that after establishing the target object chart, further data calculations can be performed based on the overall market object dataset and the target object dataset to improve the target object chart. In this embodiment, it also includes calculating the target group index (Target Group Index, TGI) in each dimension. TGI = [the proportion of the group with a certain feature in the target object group / the proportion of the group with the same feature in the overall market object group] * the standard number 100. For example, among the target objects obtained based on the screening conditions, the proportion of male objects is 50%, and the proportion of male objects in the overall market objects is 20%. Then TGI = (0.5 / 0.2) * 100 = 250, indicating that the male objects under this screening condition are more prominent and are more likely to be the target consumers of this advertisement content.

[0128] By calculating the target group index and presenting it in the target object chart, it is beneficial for advertisers to identify potential target consumers who are most likely to become sales sources, that is, the target market, and also understand the characteristics and needs of the target market, so as to optimize advertising strategies, improve advertising effects and return on investment. For example, if the TGI shows that the target object has a high influence among the media audiences in a certain region, then the advertiser can increase the advertising investment in that region to obtain better effects.

[0129] 309. Determine the play metrics for each media content respectively according to the media play conditions of the overall market objects and target objects in the media platform.

[0130] 310. Based on each play metric, sort the media content to obtain the ranking results of each media content under each play metric.

[0131] 311. Obtain the weights of each play metric.

[0132] 312. Combine the weights with the ranking results to obtain the value scores of each media content, and the value scores are used to indicate the degree of adaptation of the media content to the target advertisement.

[0133] It can be understood that steps 309 to 312 correspond to step 204 in the above Figure 2 corresponding to the step 204 in the embodiment.

[0134] In this embodiment, to determine the degree of adaptation of the media content to the target advertisement, specifically, it is to determine the value scores of each media content for the target advertisement, and the value scores are determined based on multiple play metrics.

[0135] After determining the overall market objects and target objects, first determine the play metrics corresponding to each media content respectively according to the media play conditions of these objects in the media platform. Specifically, the play metrics include the target access volume of the target object, the concentration of the target access volume in the total access volume, the concentration of the target completion volume in the total completion volume, and the completion rate preference. The above metrics correspond to the target start-play UV, penetration TA concentration, completion TA concentration, and completion rate TGI of the target object respectively. That is, the target completion volume indicates the completion volume of the target object, the completion rate preference indicates the ratio of the target completion rate to the total completion rate, the target completion rate indicates the ratio of the target completion volume to the target access volume, and the total completion rate indicates the ratio of the total completion volume to the total access volume. It can be understood that the play metrics are not limited to the above four metrics and can be replaced with various other effective content interest metrics. For example, replace the target start-play UV of the target object with the play duration of the target object, or replace the start-play UV with the effective (such as the play duration is greater than 10 seconds) play UV, etc.

[0136] Based on each playback metric, sorting the media content can obtain the ranking results of each media content under each playback metric. That is, according to different playback metrics, multiple media contents can be sorted according to their performance to determine their advantages and disadvantages in each metric.

[0137] For example, assume there are two media contents A and B, and their performances in playback metrics such as the number of visits, completion rate, and TA concentration are as follows:

[0138] The target number of visits of A is 26.09 million, the penetration TA concentration is 4.12%, and the completion TA concentration is 6.86%;

[0139] The target number of visits of B is 18.62 million, the penetration TA concentration is 5.67%, and the completion TA concentration is 4.58%.

[0140] Based on these metrics, A and B can be sorted. For example, sorted by the penetration TA concentration, the penetration TA concentration of B is higher than that of A, so B is ranked before A; sorted by the completion TA concentration, the completion TA concentration of A is higher than that of B, so A is ranked before B. In this way, the ranking results of each media content under each playback metric can be obtained.

[0141] Obtaining the weights of each playback metric means that the relative importance of each playback metric can be configured, and the weights of each playback metric are determined according to the configuration results. These weights can be pre-determined by the operators of the media platform or can be customized based on the advertising content. Based on the technical experience in this field, the weights of the target object's start-of-play UV ranking, penetration TA concentration ranking, completion TA concentration ranking, and completion rate TGI ranking can be configured as 50%, 20%, 20%, and 10% respectively.

[0142] Combining the weights with the ranking results can obtain the scores for the comprehensive evaluation of each media content. This score can be used to indicate the degree of fit between the media content and the target advertisement. For example, if the weight of the target number of visits is high, then the media content with a high target number of visits will obtain a higher score; if the weight of the completion rate TGI is low, then the media content with a low completion rate TGI will not be greatly affected.

[0143] Furthermore, the corresponding degrees of fit can be determined based on the category and / or theme of the media content:

[0144] In a possible implementation method, it specifically includes:

[0145] According to the playback situation, output the degrees of fit between multiple media categories and the target advertisement. Each media category includes multiple media themes, and each media theme includes multiple media contents;

[0146] In response to a selection operation for a specified category, output the adaptation degrees of multiple media themes under the specified category and a target advertisement, where the specified category belongs to multiple media categories;

[0147] In response to a selection operation for a specified theme, determine the adaptation degrees of each media content under the specified theme and a target advertisement, where the specified theme belongs to multiple media themes.

[0148] In this embodiment, according to the playback situation, the adaptation degrees of media categories, media themes, and media content and the target advertisement can be determined in sequence, helping the advertiser to have a more comprehensive understanding of the media content, layout the advertising placement target in multiple aspects, and improve the advertising placement strategy.

[0149] In another possible implementation method, it specifically includes:

[0150] According to the playback situation, output the adaptation degrees of multiple media categories and a target advertisement;

[0151] In response to a selection operation for a specified category, determine the adaptation degrees of each media content under the specified category and a target advertisement.

[0152] Those skilled in the art can configure the corresponding media content selection method according to actual needs.

[0153] In one possible implementation method, after step 301, it further includes:

[0154] 313. Screen out the viewer objects who have accessed the specified content from the overall market object dataset, where the specified content belongs to multiple media content;

[0155] 314. Based on the click-through rates of the viewer objects on multiple advertisement contents, determine the degree of interest of the viewer objects in each advertisement content;

[0156] It can be understood that for the specified media content, this embodiment also provides a method for attracting investment, which is used to select matching advertisement content for the specified media, assist in sales to find advertisers, so as to achieve full sales of various media content.

[0157] First, screen out those objects who have ever accessed the specified content from the overall market object dataset. The overall market object dataset usually contains various behavior data of a large number of objects, including access, viewing, clicking, etc. By screening out the objects who have accessed the specified content, it can be determined which objects are interested in this content, so as to more accurately understand the needs and preferences of this part of the objects.

[0158] Then, by observing the click-through rates of the audience on multiple advertising contents, the degree of their interest in each advertising content is determined. The click-through rate is one of the important indicators to measure the degree of an object's interest in content or an advertisement. By analyzing the click-through rate, it can be understood which advertising contents are more popular among the audience, thereby providing a basis for subsequent advertising strategies.

[0159] Finally, according to the media playback situations of the overall market objects and the audience objects in the media platform, the adaptation degrees of multiple media contents for a specified advertisement are determined, thereby assisting in finding more suitable advertising contents.

[0160] In a possible implementation method, for media contents that have not yet been broadcast, it is difficult to obtain the corresponding audience objects for such media contents, and it is difficult to find suitable advertisers. Therefore, the reservation objects of the media contents can be used as the audience objects, and the media contents and advertising contents are matched through the methods of step 313 to step 315 above to assist in business opportunity negotiation before broadcasting.

[0161] For the convenience of understanding, a content recommendation method applied to a content recommendation tool will be introduced below. The following method takes a video playback platform as an example. It can be understood that the tools, the content of the application examples, and the users and populations in their corresponding drawings are equivalent to the objects and the object groups composed of a certain type of object in the above text.

[0162] Tool 1. A media content recommendation tool based on advertising content, please refer to Figures 4a to 4f :

[0163] 1) After obtaining the overall market object dataset of the video playback platform, according to the advertising content, the screening conditions for each dimension are determined. Please refer to Figure 4a , it is determined that the advertising type of the target advertisement is an automobile advertisement, and the industry label it belongs to is the automobile industry; among the screening conditions for each dimension, the age range is determined to be 18 - 34 years old, the city levels include first-tier cities, second-tier cities, and new first-tier cities, and other dimensions are not limited.

[0164] 2) According to the above screening conditions, the number of target objects obtained and the target object chart. Please refer to Figure 4b , the number of target objects is 11434149, and the target object chart includes, for each dimension, the proportion of target objects corresponding to each object feature and the TGI.

[0165] 3) According to the media playback situations of the overall market objects and the target objects in the media platform, the playback indicators for each media asset category (media resource category) are determined. Please refer to Figure 4c , including the playback indicators for each media asset category, and for each media asset category, according to the weight distribution of the preset playback indicators, the category score of each media asset category is calculated to obtain the adaptation degree of each media asset category to the target advertisement.

[0166] 4) After selecting TV dramas in the media asset category, determine the play metrics for the IPs under this category. Please refer to Figure 4d , including the play metrics for each IP under the TV drama category, and for each IP, perform weight allocation according to the preset play metrics, calculate the category scores for each IP, and obtain the adaptation degree of each IP to the target advertisement.

[0167] 5) After selecting TV dramas in the media asset category, it is also possible to determine the play metrics for each theme under this category. Please refer to Figure 4e , including the play metrics for each theme under the TV drama category, and for each theme, perform weight allocation according to the preset play metrics, calculate the category scores for each theme, and obtain the adaptation degree of each theme to the target advertisement.

[0168] 6) It supports sorting by a single play metric and provides a data download function, such as Figure 4f .

[0169] Tool 2: Advertising content recommendation tool based on media content. Please refer to Figures 5a to 5c :

[0170] 1) It is possible to select media IPs based on the category, type, and content name, such as Figure 5a shown, select the media IP with the category of variety show, content type of IP, and content name of "Battle XXX Season 2".

[0171] 2) Obtain the number of viewer objects who have accessed the IP and the viewer object chart based on the media IP. Please refer to Figure 5b , the number of viewer objects is 5367777, and the viewer object chart includes the proportion and TGI of viewer objects corresponding to each object feature under each dimension.

[0172] 3) Based on the click-through rates of multiple advertising contents by the viewer objects, it is possible to determine the degree of interest of the viewer objects in different advertising categories, such as Figure 5c shown, including the click-through rates of the viewer objects for various categories of advertisements, and the ratio of the click-through rate of the viewer objects to the click-through rate of the overall market objects.

[0173] Tool 3: Advertising content recommendation tool based on unplayed IPs. Please refer to 6a and Figure 6b :

[0174] 1) Select the category and IP, such as Figure 6a shown;

[0175] 2) Obtain the number of reservation objects and the reservation object chart based on the IP per day. Please refer to Figure 6b, the number of reserved objects is 4,043,387, and the reserved object chart includes, for each dimension, the proportion of reserved objects corresponding to each object feature and the TGI.

[0176] 3) According to the click-through rates of multiple advertising contents by reserved objects, the degree of interest of reserved objects in different advertising categories can be determined. It can be understood that this step is similar to step 3) in Scenario 2.

[0177] The content recommendation method provided by the embodiments of the present application calculates the adaptation degree with advertising contents respectively from the aspects of media category, media theme and media content based on multiple media playback metrics, which is conducive to realizing efficient and accurate recommendation of advertising contents, and at the same time realizing more efficient sale of media content resources, reaching and conversion of advertiser target objects.

[0178] The following provides several application scenario examples based on the above content recommendation method:

[0179] Scenario 1, Home Furnishing Industry:

[0180] Background: For the advertising business opportunities of furniture manufacturers, advertisers have reported three main consumer groups. Combining the previous analysis conclusions, advertising placement is greatly affected by the playback VV. Most of the relevant recommended variety shows are the top content on the platform, lacking the ability of differential recommendation based on the content preferences of different industry objects (differential recommendation not only helps in customized planning for industry investment promotion, but also helps in the inventory sale of non-top content). Therefore, each variety show IP is analyzed from dimensions such as TA concentration (based on penetrated and fully watched objects).

[0181] Target objects: Can be screened based on multi-dimensional features, specifically including: 1. The group of newly graduated college students; 2. The group of people aged 25 - 45 with houses; 3. The high-consumption group aged 45 - 55. The corresponding object chart selection logic is as Figure 7a described.

[0182] Analysis conclusion: Group 1 is more inclined to young-oriented variety shows, Group 2 as a whole is more inclined to comedy-oriented variety shows, and Group 3 is more inclined to comedy-oriented and music-oriented variety shows. And the recommendation results reduce the influence of playback VV and realize differential recommendation based on the content preferences of different groups.

[0183] For Group 1, the recommended results of variety show IPs are as Figure 7b shown. The ratings are determined based on various playback metrics.

[0184] Scenario 2, Automobile Industry:

[0185] Background: For the advertising business opportunities of a certain automobile manufacturer, AA Automobile, evaluate its object characteristics and content consumption preferences on the video platform with competing objects (BB Automobile and CC Automobile), empower the guidance of pre-sale content resources and the formulation of marketing plans, and promote the implementation of business opportunities.

[0186] Target objects: Conduct target object mining (users of AA cars and competing models).

[0187] Analysis conclusion: As Figure 8 shown, for the AA car population, the category interests mainly focus on sports, culture and history, and documentaries, and output the top 30 IPs and themes in each category; for the competing model population, the top three category interests are also sports, culture and history, and documentaries.

[0188] Scenario 3, gaming industry:

[0189] Background: For the advertising business opportunities of a certain game manufacturer, it is hoped to evaluate the object characteristics and content consumption preferences of the target customers on the video platform, promote the implementation of business opportunities, empower advertisers to quickly expand the domestic market, cultivate user minds, and quickly connect with the international version.

[0190] Target objects: Conduct target population mining (users of the same type of games).

[0191] Analysis conclusion: As Figure 9 shown, based on the start-up UV of the target population for each category, the main consumption categories of the target population are anime, TV dramas and movies, and output the corresponding IP and theme recommendations for each category.

[0192] Scenario 4, financial industry:

[0193] Background: Based on the substantial increase in residents' deposits and savings in a certain year, the business side evaluates that there are great growth opportunities in the financial industry. It is hoped to evaluate the population coverage, population characteristic distribution, content consumption preferences and resource position distribution of the financial industry on the video platform, and empower the business side to promote industry large customer business opportunities.

[0194] Target objects: Conduct target population mining (determine the target population based on the click situation of financial content and financial advertisements).

[0195] Analysis conclusion: As Figure 10 shown, by synthesizing the start-up UV of the overall market population, the penetration TA concentration, the completion TA concentration and the TA completion rate, output the IP interest preferences of the three target populations in the top categories.

[0196] Scenario 5, tourism industry:

[0197] Background: With the industry recovery, it is hoped to inventory the content consumption interests of the target population in the tourism industry, empower the formulation of scenario marketing plans for industry customers, and strive for more customer budgets.

[0198] Target Objects: The advertisers have a clear understanding and can screen based on multi-dimensional features, specifically including: 1. Parent-child population (parenting status 0 - 12 years old, marital status married, age 25 - 45 years old, limited to permanent residence cities); 2. Young population (18 - 29 years old, limited to permanent residence cities)

[0199] Analysis Conclusion: As Figure 11 shown, based on the play data by category, IP interest analysis and recommendation are prioritized for the five categories of TV dramas, movies, variety shows, animations, and children's programs. By integrating the start-up UV of the overall market population, the penetration TA concentration, the completion TA concentration, and the TA completion rate, the IP interest preferences of the parent-child population and the young population are output.

[0200] The content recommendation device in this application will be described in detail below. Please refer to Figure 12 . Figure 12 FIG. 1200 is a schematic diagram of an embodiment of the content recommendation device 1200 in this application. The content recommendation device 1200 includes:

[0201] An acquisition module 1201, configured to acquire an overall market object data set of a media platform. The media platform includes multiple media contents. The overall market object data set includes object data of multiple overall market objects, and each object data includes object characteristics of multiple dimensions;

[0202] A condition determination module 1202, configured to determine the screening conditions for each dimension based on the advertisement type of the target advertisement;

[0203] A screening module 1203, configured to screen out multiple target objects that meet the screening conditions from the overall market object data set;

[0204] An adaptation module 1204, configured to determine the adaptation degree of multiple media contents to the target advertisement respectively according to the media play conditions of the overall market objects and the target objects in the media platform.

[0205] The content recommendation device provided in the embodiment of this application acquires the object characteristics of the overall market objects in each dimension of the media platform, then screens the overall market objects based on the advertisement type of the target advertisement to obtain target objects, and then determines the adaptation degree of the media content to the advertisement content based on the media play conditions of the target objects and the overall market objects. Advertisers can use this adaptation degree as a reference for advertising placement. The content recommendation method provided in this embodiment comprehensively considers the advertisement content and the media content, rather than only considering the number of plays of the media content, which is beneficial to realizing more efficient sales of media content resources and improving the conversion rate of advertisement content.

[0206] In a possible implementation method, it further includes:

[0207] A chart creation module 1205 is configured to establish a target object dataset based on multiple said target objects, where the target object dataset includes the object data of multiple said target objects; and establish a target object chart according to the target object dataset, where the target object chart indicates the distribution of the target objects in multiple said dimensions.

[0208] In a possible implementation method, it further includes:

[0209] The chart creation module 1205 is further configured to determine a target group index in multiple said dimensions based on the overall market object dataset and the target object dataset, where the target group index indicates the ratio of a first distribution to a second distribution, the first distribution indicates the distribution of the target objects in multiple said dimensions, and the second distribution indicates the distribution of the overall market objects in multiple said dimensions; and display the target group index in the target object chart.

[0210] In a possible implementation method, the condition determination module 1202 is specifically configured to determine a corresponding consumer object based on the advertisement type of the target advertisement; and determine the screening conditions based on the consumer object.

[0211] In a possible implementation method, the adaptation module 1204 is specifically configured to determine a playback metric for each said media content respectively according to the media playback situations of the overall market objects and the target objects in the media platform; sort the media content based on each said playback metric to obtain a ranking result of each said media content under each said playback metric; obtain the weight of each said playback metric; and combine the weight with the ranking result to obtain a value score for each said media content, where the value score is used to indicate the degree of adaptation of the media content to the target advertisement.

[0212] In a possible implementation method, the adaptation module 1204 is specifically configured to output the degree of adaptation of multiple media categories to the target advertisement according to the playback situation, where each said media category includes multiple media themes, and each said media theme includes multiple said media content; in response to a selection operation on a specified category, output the degree of adaptation of multiple said media themes under the specified category to the target advertisement, where the specified category belongs to multiple said media categories; and in response to a selection operation on a specified theme, determine the degree of adaptation of each said media content under the specified theme to the target advertisement, where the specified theme belongs to multiple said media themes.

[0213] In a possible implementation method, it further includes:

[0214] The media content analysis module filters out the viewer objects that have accessed the specified content from the large-scale object dataset, and the specified content belongs to multiple pieces of the media content; based on the click-through rates of the viewer objects on multiple advertisement contents, the degree of interest of the viewer objects in each advertisement content is determined.

[0215] It can be understood that the content recommendation device provided in the embodiments of the present application corresponds to the content recommendation method in the above Figure 2 and Figure 3 corresponding embodiments. For specific descriptions, please refer to the above, and details will not be elaborated here.

[0216] The embodiments of the present application further provide a computer device, including a memory and a processor. The memory stores instructions, and when the instructions run on the processor, the steps of the content recommendation method as described in Figure 2 or Figure 3 are executed.

[0217] This computer device may be Figure 1 the terminal device or server shown in the figure. In the embodiments of the present application, taking the computer device as a server as an example, the structure of this computing device will be introduced. Please refer to Figure 13 , Figure 13 which is a schematic structural diagram of a server provided in the embodiments of the present application. The server 300 may vary greatly due to different configurations or performances, and may include one or more central processing units (CPUs) 322 (for example, one or more processors) and a memory 332, and one or more storage media 330 (for example, one or more mass storage devices) for storing application programs 342 or data 344. Among them, the memory 332 and the storage media 330 may be short-term storage or persistent storage. The programs stored in the storage media 330 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the server. Further, the central processor 322 may be configured to communicate with the storage media 330 and execute a series of instruction operations in the storage media 330 on the server 300.

[0218] The server 300 may further include one or more power supplies 326, one or more wired or wireless network interfaces 350, one or more input / output interfaces 358, and / or one or more operating systems 341, such as Windows Server TM , Mac OS X TM , Unix TM , Linux TM , FreeBSD TM and so on.

[0219] In the above embodiments, the steps executed by the server can be based on the Figure 13 server structure shown.

[0220] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, device, and unit can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0221] In the embodiments of the present application, the term "module" or "unit" refers to a computer program with a predetermined function or a part of a computer program, which works together with other related parts to achieve a predetermined goal, and can be fully or partially implemented by using software, hardware (such as a processing circuit or a memory), or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be a part of an overall module or unit that includes the function of that module or unit.

[0222] In several embodiments provided by the present application, it should be understood that the disclosed system, device, and method can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces, and the indirect coupling or communication connection of the device or unit can be in electrical, mechanical, or other forms.

[0223] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0224] In addition, the functional units in the various embodiments of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0225] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of this application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0226] The above, the above embodiments are only used to illustrate the technical solution of this application and are not intended to limit it; although this application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of various embodiments of this application.

Claims

1. A content recommendation method, characterized in that, Including: Obtain a large-scale object dataset of a media platform, where the media platform includes a plurality of media contents, the large-scale object dataset includes object data of a plurality of large-scale objects, and each piece of object data includes object features in multiple dimensions; Determine the screening conditions for each dimension based on the advertisement type of the target advertisement; Screen out a plurality of target objects that meet the screening conditions from the large-scale object dataset; Determine the adaptation degree between the plurality of media contents and the target advertisement respectively according to the media playback situations of the large-scale objects and the target objects in the media platform.

2. The method according to claim 1, wherein After screening out a plurality of target objects that meet the screening conditions from the large-scale object dataset, it further includes: Establish a target object dataset based on the plurality of target objects, where the target object dataset includes the object data of the plurality of target objects; Establish a target object chart according to the target object dataset, and the target object chart indicates the distribution of the target objects in multiple dimensions.

3. The method according to claim 2, wherein After establishing the target object chart according to the target object dataset, it further includes: Based on the large-scale object dataset and the target object dataset, determine the target group index in multiple dimensions, where the target group index indicates the ratio of the first distribution to the second distribution, the first distribution indicates the distribution of the target objects in multiple dimensions, and the second distribution indicates the distribution of the large-scale objects in multiple dimensions; Display the target group index in the target object chart.

4. The method according to claim 1, wherein The determining the screening conditions for each dimension based on the advertisement type of the target advertisement includes: Determine the corresponding consumer objects based on the advertisement type of the target advertisement; Determine the screening conditions based on the consumer objects.

5. The method according to claim 1, wherein The determining the adaptation degree between the plurality of media contents and the target advertisement respectively according to the media playback situations of the large-scale objects and the target objects in the media platform includes: Determine the playback metrics of each media content respectively according to the media playback situations of the large-scale objects and the target objects in the media platform; Sort the media contents based on each playback metric to obtain the ranking results of each media content under each playback metric; Obtain the weight of each playback metric; Combine the weight with the ranking results to obtain the value score of each media content, and the value score is used to indicate the adaptation degree between the media content and the target advertisement.

6. The method according to claim 1, wherein The determining the adaptation degree between the plurality of media contents and the target advertisement respectively according to the media playback situations of the large-scale objects and the target objects in the media platform includes: Output the adaptation degree between a plurality of media categories and the target advertisement according to the playback situation, where each media category includes a plurality of media themes, and each media theme includes a plurality of the media contents; In response to a selection operation on a specified category, output the adaptation degree between a plurality of the media themes under the specified category and the target advertisement, where the specified category belongs to the plurality of media categories; In response to a selection operation on a specified theme, determine the degree of suitability of each of the media contents under the specified theme for the target advertisement, where the specified theme belongs to multiple media themes.

7. The method according to claim 1, wherein After obtaining the overall market object dataset of the media platform, it further includes: Screen out the viewer objects that have accessed the specified content from the overall market object dataset, where the specified content belongs to multiple media contents; Based on the click-through rates of the viewer objects on multiple advertisement contents, determine the degree of interest of the viewer objects in each of the advertisement contents.

8. A content recommendation device, characterized in that, It includes: An acquisition module, configured to acquire the overall market object dataset of the media platform, where the media platform includes multiple media contents, the overall market object dataset includes the object data of multiple overall market objects, and each object data includes object features in multiple dimensions; A condition determination module, configured to determine the screening conditions for each dimension based on the advertisement type of the target advertisement; A screening module, configured to screen out multiple target objects that meet the screening conditions from the overall market object dataset; An adaptation module, configured to determine the degree of suitability of multiple media contents for the target advertisement respectively according to the media playback situations of the overall market objects and the target objects in the media platform.

9. A computer device, characterized in that, It includes: A memory and a processor; The memory stores instructions, and when the instructions run on the processor, execute the content recommendation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, including instructions, when it runs on a computer device, causes the computer device to execute the content recommendation method according to any one of claims 1 to 7.

Images