Media content recommendation method and device, electronic equipment and storage medium

Abstract

The invention relates to a media content recommendation method and device, electronic equipment and a storage medium. The media content recommendation method comprises the following steps: acquiring a content value score of first media content, and acquiring related data of the first media content; and determining first recommendation data according to the content value score of the first media content and the related data of the first media content. Second recommendation data is determined according to the first recommendation data and a received control instruction, the second recommendation data comprises the push priority of the second media content, the control instruction is used for adjusting the push priority of the first media content, and the second media content is different from the first media content and comprises multiple media contents. And determining and pushing the target media content according to the pushing priority of the second media content and a preset priority threshold. By means of the method and device, it is guaranteed that content recommendation is efficiently executed through an algorithm, the personalized requirement of a user is met, and meanwhile new hot content recommendation can be conducted through manual intervention.

Description

Technical Field

[0001] The present disclosure relates to the field of smart Internet television, and in particular to a media content recommendation method, device, electronic device and storage medium. Background Art

[0002] The smart Internet TV industry has gone through more than ten years of development since its birth. During this process, the recommendation model of TV content operation has also experienced a transformation from the initial manual operation to personalized algorithm recommendation.

[0003] However, the traditional manual recommendation method has the problems of low recommendation efficiency and inability to meet the personalized requirements of users. The algorithm recommendation method is insensitive to new and popular content and lacks flexibility. Summary of the invention

[0004] In order to overcome the problems existing in the related art, the present disclosure provides a media content recommendation method, device, electronic device and storage medium.

[0005] According to a first aspect of an embodiment of the present disclosure, a content recommendation method is provided, comprising: obtaining a content value score of a first media content and obtaining relevant data of the first media content, the first media content including multiple media contents, the content value score representing the comprehensive quality of the media content, and the relevant data representing the popularity of the media content; determining first recommendation data based on the content value score of the first media content and the relevant data of the first media content, the first recommendation data including a push priority of the first media content; determining second recommendation data based on the first recommendation data and a received control instruction, the second recommendation data including a push priority of a second media content, the control instruction being used to adjust the push priority of the first media content, the second media content being different from the first media content and including multiple media contents; determining and pushing target media content based on the push priority of the second media content and a preset priority threshold.

[0006] In one embodiment, obtaining the content value score of the first media content includes: determining the content tag, traffic data, and popularity data of the media content for each of the multiple media contents in the first media content; determining a first value score of the media content based on the content tag, the first value score representing the content quality of the media content, and determining a second value score of the media content based on the traffic data, the second value score representing the platform traffic occupied by the media content, and determining a third value score of the media content based on the popularity data, the second value score representing the online popularity of the media content; and determining the sum of the first value score, the second value score, and the third value score as the content value score of the media content.

[0007] In one implementation, the data related to the first media content includes one or more of the following data: playback penetration rate of the media content, payment conversion rate of the media content, or user payment data when the media content is pushed a preset number of times.

[0008] In one embodiment, the method further includes: based on the control instruction, setting a first preset media content and setting a priority for the first preset media content, wherein the first preset media content is different from the first media content; and / or removing recommendation data corresponding to a second preset media content in the first recommendation data, wherein the first media content includes the second preset media content.

[0009] In one embodiment, the pushing of target media content includes: determining target traffic data for a target time period, wherein the target time period is a next time period adjacent to the current time period; determining a target recommendation number corresponding to the target traffic data based on a correspondence between traffic data and push times; determining target push data based on current push data, a first number of currently online first-category users, a second number of currently online second-category users, the target recommendation number, and a first preset method, wherein the current push data includes a first push probability corresponding to the first-category users and a second push probability corresponding to the second-category users, the target push data includes a first target push probability corresponding to the first-category users and a second push probability corresponding to the second-category users, the first-category users are users who are interested in the target media content, and the second-category users are users who are not interested in the media content; and according to the target push data, pushing the target media content to the first-category users and the second-category users.

[0010] In one embodiment, the target push data is determined based on the current push data, the first number of the first category of users currently online, the second number of the second category of users currently online, the target push number and the first preset method, including: determining a first product of a first push probability and the first number, and determining a second product of a second push probability and the second number; in response to the sum of the first product and the second product being equal to the target push number, determining the first push probability as the first target push probability, and determining the second push probability as the second target push probability; in response to the sum of the first product and the second product being less than the target push number, increasing the first push probability, And determine the third product of the increased first push probability and the first number, until the sum of the third product and the second product is equal to the target push number, determine the increased first push probability as the first target push probability, and determine the second push probability as the second target push probability; in response to the sum of the first product and the second product being greater than the target push number, reduce the second push probability, and determine the fourth product of the reduced second push probability and the second number, until the sum of the fourth product and the first product is equal to the target push number, determine the reduced second push probability as the second target push probability, and determine the first push probability as the first target push probability.

[0011] In one embodiment, the target flow data for a target time period is determined, including: performing data fitting for the target time period based on first historical flow data through a preset model to obtain fitted flow data corresponding to the target time period, wherein the first historical flow data includes flow data corresponding to different time periods within a first cycle from the target time period; determining second historical flow data for the historical time period, and performing data fitting for the historical time period through a preset model to obtain historical fitted flow data corresponding to the historical time period, wherein the historical time period is a time period between 0:00 on the day and a preset time interval before the current time period; adjusting the fitted flow data according to the difference ratio between the second historical flow data and the historical fitted flow data to obtain the target flow data.

[0012] According to a second aspect of an embodiment of the present disclosure, a content recommendation device is provided, comprising: an acquisition unit, configured to acquire a content value score of a first media content and acquire relevant data of the first media content, wherein the first media content includes multiple media contents, the content value score represents the comprehensive quality of the media content, and the relevant data represents the popularity of the media content; a determination unit, configured to determine first recommendation data according to the content value score of the first media content and the relevant data of the first media content, wherein the first recommendation data includes a push priority of the first media content; a processing unit, configured to determine second recommendation data according to the first recommendation data and a received control instruction, wherein the second recommendation data includes a push priority of a second media content, wherein the control instruction is used to adjust the push priority of the first media content, wherein the second media content is different from the first media content and includes multiple media contents; and a push unit, configured to determine and push target media content according to the push priority of the second media content and a preset priority threshold.

[0013] In one embodiment, the acquisition unit acquires the content value score of the first media content in the following manner: for each media content in multiple media contents in the first media content, determine the content tag, traffic data and popularity data of the media content; determine the first value score of the media content according to the content tag, the first value score represents the content quality of the media content, and determine the second value score of the media content according to the traffic data, the second value score represents the platform traffic occupied by the media content, and determine the third value score of the media content according to the popularity data, the second value score represents the online popularity of the media content; determine the sum of the first value score, the second value score and the third value score as the content value score of the media content.

[0014] In one implementation, the data related to the first media content includes one or more of the following data: playback penetration rate of the media content, payment conversion rate of the media content, or user payment data when the media content is pushed a preset number of times.

[0015] In one embodiment, the processing unit is also used to: based on the control instruction, set a first preset media content and set a priority for the first preset media content, the first preset media content is different from the first media content; and / or remove the recommendation data corresponding to the second preset media content in the first recommendation data, the first media content includes the second preset media content.

[0016] In one embodiment, the push unit pushes the target media content in the following manner: determining target traffic data for a target time period, wherein the target time period is a next time period adjacent to the current time period; determining a target recommendation number corresponding to the target traffic data based on a correspondence between traffic data and push times; determining target push data based on current push data, a first number of currently online first-category users, a second number of currently online second-category users, the target recommendation number, and a first preset method, wherein the current push data includes a first push probability corresponding to the first-category users and a second push probability corresponding to the second-category users, the target push data includes a first target push probability corresponding to the first-category users and a second push probability corresponding to the second-category users, the first-category users are users who are interested in the target media content, and the second-category users are users who are not interested in the media content; and pushing the target media content to the first-category users and the second-category users based on the target push data.

[0017] In one embodiment, the push unit determines the target push data according to the current push data, the first number of the first category of users currently online, the second number of the second category of users currently online, the target push number and a first preset method in the following manner: determining a first product of a first push probability and the first number, and determining a second product of a second push probability and the second number; in response to the sum of the first product and the second product being equal to the target push number, determining the first push probability as the first target push probability, and determining the second push probability as the second target push probability; in response to the sum of the first product and the second product being less than the target push number, increasing the first push probability. The push probability is determined, and a third product of the increased first push probability and the first number is determined, until the sum of the third product and the second product is equal to the target push number, the increased first push probability is determined as the first target push probability, and the second push probability is determined as the second target push probability; in response to the sum of the first product and the second product being greater than the target push number, the second push probability is reduced, and a fourth product of the reduced second push probability and the second number is determined, until the sum of the fourth product and the first product is equal to the target push number, the reduced second push probability is determined as the second target push probability, and the first push probability is determined as the first target push probability.

[0018] In one embodiment, the push unit determines the target flow data of the target time period in the following manner: through a preset model, based on the first historical flow data, data fitting is performed for the target time period to obtain the fitted flow data corresponding to the target time period, the first historical flow data includes flow data corresponding to different time periods within a first cycle from the target time period; second historical flow data of the historical time period is determined, and data fitting is performed on the historical time period through a preset model to obtain historical fitted flow data corresponding to the historical time period, the historical time period is the time period between 0:00 on the day and a preset time interval before the current time period; according to the difference ratio between the second historical flow data and the historical fitted flow data, the fitted flow data is adjusted to obtain the target flow data.

[0019] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor executable instructions; wherein the processor is configured to: execute the media content recommendation method described in the first aspect or any one of the implementations of the first aspect.

[0020] According to a fourth aspect of an embodiment of the present disclosure, a storage medium is provided, in which instructions are stored. When the instructions in the storage medium are executed by a processor, the processor is enabled to execute the media content recommendation method described in the first aspect or any one of the embodiments of the first aspect.

[0021] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: for the first media content, a content value score representing the first quality of the media content is obtained, and relevant data representing the popularity of the media content is obtained. Based on the content value score and relevant data of the first media content, first recommendation data is determined, and the first recommendation data includes the push priority of the first media content. The push priority of the first media content is adjusted according to the received control instruction, and second recommendation data is determined, and the second recommendation data includes the push priority of the second media content, and the second media content is different from the first media content and includes multiple media contents. According to the push priority of the second media content and the preset priority threshold, the target media content is determined and pushed. Through the present disclosure, while ensuring that content recommendation is performed efficiently through the algorithm and meets the personalized requirements of users, new hot content can be recommended through manual intervention.

[0022] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0024] Figure 1 The figure is a flow chart of a method for recommending media content according to an exemplary embodiment.

[0025] Figure 2 The present invention is a flowchart of a method for obtaining content value points of first media content according to an exemplary embodiment.

[0026] Figure 3 The figure is a schematic diagram showing a method for obtaining content value points of media content according to an exemplary embodiment of the present disclosure.

[0027] Figure 4 The figure is a schematic diagram showing a method for obtaining content value points of media content according to an exemplary embodiment of the present disclosure.

[0028] Figure 5 The invention is a flowchart of a method for setting first media content according to an exemplary embodiment.

[0029] Figure 6 It is a schematic diagram of a method for setting recommended content according to a control instruction according to an exemplary embodiment of the present disclosure.

[0030] Figure 7 The present invention is a flowchart of a method for pushing target media content according to an exemplary embodiment.

[0031] Figure 8 The present invention is a flowchart showing a method for determining target push data according to an exemplary embodiment.

[0032] Fig. 9 The invention is a flowchart showing a method for determining target flow data in a target time period according to an exemplary embodiment.

[0033] Fig.10 It is a schematic diagram showing a method for determining target flow data in a target time period according to an exemplary embodiment of the present disclosure.

[0034] Fig.11 It is a schematic diagram showing historical traffic data according to an exemplary embodiment of the present disclosure.

[0035] Fig.12 It is a schematic diagram showing fitting data and actual data according to an exemplary embodiment of the present disclosure.

[0036] Fig.13 It is a schematic diagram of a media content recommendation method according to an exemplary embodiment of the present disclosure.

[0037] Fig.14 It is a schematic diagram of a media content recommendation architecture according to an exemplary embodiment of the present disclosure.

[0038] Fig.15 The figure is a block diagram of a media content recommendation device according to an exemplary embodiment.

[0039] Fig.16 It is a block diagram of a device for media content recommendation according to an exemplary embodiment.

[0040] Fig.17 It is a block diagram of a device for media content recommendation according to an exemplary embodiment. DETAILED DESCRIPTION

[0041] Here, exemplary embodiments will be described in detail, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present disclosure.

[0042] The media content recommendation method provided by the embodiments of the present disclosure is applied to a scenario where media content is recommended by combining an algorithmic approach and a manual approach.

[0043] The smart Internet TV industry has gone through more than ten years of development since its birth. During this process, the recommendation model of TV content operation has undergone a transformation from the initial manual operation recommendation to personalized algorithm recommendation. Algorithmic recommendation is significantly better than traditional manual recommendation in improving recommendation efficiency, conversion rate and optimizing user experience. However, algorithm recommendation is not perfect. It faces challenges such as recommendation accuracy problems caused by insufficient recognition of the life cycle of new and hot content, and decreased conversion rate caused by overexposure of top content.

[0044] In the related technology, for the manual recommendation method, experienced editors or operators select content based on professional knowledge, which can ensure the quality and suitability of the content, and can quickly adjust the recommended content according to specific situations such as current hotspots, festivals and celebrations, which is more timely and targeted, and can set the recommended content based on the overall inclination of the operator. It has the characteristics of strong professionalism and high flexibility, and better reflects and conveys specific values ​​and cultural orientations. However, manual recommendation requires a lot of manpower and material resources, and the processing speed is limited. It is difficult to cope with the needs of large-scale users, and is affected by personal preferences and prejudices. It cannot fully meet the personalized needs of users, and due to human resource limitations, manual recommendation is difficult to achieve comprehensive coverage of massive content. Therefore, manual recommendation has the problems of low efficiency, strong subjectivity and limited coverage.

[0045] In the related technology, the algorithm recommendation method can process a large amount of data, provide users with fast and personalized content recommendations, and make recommendations based on user behavior and preferences, which can more accurately meet user needs, reduce manual intervention, and reduce operating costs. It has the characteristics of high accuracy and high efficiency. However, the algorithm recommendation method is difficult to deal with emergencies and special dates / situations, which may result in the recommended content not meeting the actual needs of users, and is often slow to respond to new and hot content, affecting the recommendation effect. There are problems such as lack of flexibility and insufficient recognition of new content.

[0046] In view of this, the present disclosure proposes a method for media content recommendation, which obtains, for a first media content, a content value score representing a first quality of the media content, and obtains relevant data representing the popularity of the media content. Based on the content value score and relevant data of the first media content, first recommendation data is determined, and the first recommendation data includes a push priority of the first media content. The push priority of the first media content is adjusted according to the received control instruction, and second recommendation data is determined, and the second recommendation data includes a push priority of second media content, and the second media content is different from the first media content and includes multiple media contents. Based on the push priority of the second media content and a preset priority threshold, the target media content is determined and pushed. Through the present disclosure, while ensuring that content recommendation is performed efficiently through an algorithm and meets the personalized requirements of users, new hot content can be recommended through manual intervention.

[0047] Figure 1 FIG. 1 is a flow chart of a method for recommending media content according to an exemplary embodiment. Figure 1 As shown, the method includes steps S101 to S104.

[0048] In step S101, the content value score of the first media content is obtained, and the related data of the first media content is obtained. The first media content includes multiple media contents. The content value score represents the comprehensive quality of the media content, and the related data represents the popularity of the media content.

[0049] In step S102, first recommendation data is determined according to the content value score of the first media content and the relevant data of the first media content, where the first recommendation data includes a push priority of the first media content.

[0050] In step S103, second recommendation data is determined according to the first recommendation data and the received control instruction, the second recommendation data includes the push priority of the second media content, the control instruction is used to adjust the push priority of the first media content, and the second media content is different from the first media content and includes multiple media content.

[0051] In step S104, target media content is determined and pushed according to the push priority of the second media content and a preset priority threshold.

[0052] In the embodiment of the present disclosure, the algorithm recommendation method and the manual recommendation method are combined to recommend media content. The recommendation algorithm can calculate the value points for the existing media content (first media content, including TV series, movies, animation, variety shows, etc.) of the video platform (intelligent Internet TV platform) to obtain the content value points that characterize the comprehensive quality of the media content. And it is possible to obtain relevant data that characterizes the degree of recommendation of the media content by obtaining user-side data. It can be understood that the degree of recommendation of the media content and the comprehensive quality of the media content can both be used as the basis for recommending the media content. Therefore, after determining the content value points and relevant data of the first media content, the present disclosure determines the push priority (first recommendation data) of each media content in the first media content according to the content value points and relevant data through the recommendation algorithm. It can be understood that the recommendation algorithm has certain limitations, is insensitive to new hot content, and lacks predictability. The present disclosure can further adjust the first recommendation data obtained by the algorithm based on the control instructions sent by the user, such as adjusting the push priority of each media content in the first media content to obtain the second media content. Through the present disclosure, while ensuring that the algorithm efficiently performs content recommendation and meets the personalized requirements of users, it is possible to recommend new hot content through manual intervention.

[0053] It is understandable that the video platform contains a huge amount of media content, and when users browse media content through the video platform, the content browsed in the recommendation interface is relatively limited. Therefore, the present disclosure is provided with a preset priority threshold, and the second media content is further screened through the preset priority threshold, and the media content in the second media content whose push priority meets the priority threshold requirement is used as the target media content and recommended.

[0054] In an embodiment of the present disclosure, for a first media content, a content value score representing a first quality of the media content is obtained, and related data representing the popularity of the media content is obtained. Based on the content value score and related data of the first media content, first recommendation data is determined, and the first recommendation data includes a push priority of the first media content. Based on the received control instruction, the push priority of the first media content is adjusted, and second recommendation data is determined, and the second recommendation data includes a push priority of second media content, and the second media content is different from the first media content and includes multiple media contents. Based on the push priority of the second media content and a preset priority threshold, the target media content is determined and pushed. Through the present disclosure, an algorithm recommendation method and a manual recommendation method are combined to perform media content recommendation, and while ensuring that content recommendation is performed efficiently through the algorithm and meets the personalized requirements of users, new hot content can be recommended through manual intervention.

[0055] It is understandable that when evaluating the comprehensive quality of media content, it is necessary to evaluate the different dimensions of the media content. The present disclosure obtains the value score of each dimension of the media content for different dimensions of the media content, and then determines the content value score of the media content based on the value score of each dimension.

[0056] Figure 2 FIG. 1 is a flow chart showing a method for obtaining content value points of first media content according to an exemplary embodiment. Figure 2 As shown, the method includes steps S201 to S203.

[0057] In step S201, for each media content in the plurality of media content in the first media content, a content tag, flow data, and popularity data of the media content are determined.

[0058] In step S202, a first value score of the media content is determined based on the content tag, the first value score representing the content quality of the media content, and a second value score of the media content is determined based on the traffic data, the second value score representing the platform traffic occupied by the media content, and a third value score of the media content is determined based on the popularity data, the second value score representing the online popularity of the media content.

[0059] In step S203, the sum of the first value score, the second value score and the third value score is determined as the content value score of the media content.

[0060] In the disclosed embodiment, for the first media content, value points are obtained from multiple dimensions such as the content quality of the media content, the platform traffic occupied, and the online popularity. For the content quality dimension of the media content, the content tags of the media content (such as director, starring, type, subject matter, etc.) are obtained, and the first value score of the content quality representing the quality of the media content is determined based on the content tags of the media content. For the platform traffic dimension occupied by the media content, the traffic data of the media content (such as program exposure and program revenue) is obtained, and the second value score representing the platform traffic occupied by the media content is determined based on the traffic data. For the online popularity of the media content, the popularity data of the media content (such as program popularity and filmmaker popularity, etc.) is obtained, and the third value score representing the online popularity of the media content is determined based on the popularity data. The first value score, the second value score, and the third value score are then determined as the content value score.

[0061] In the embodiment of the present disclosure, different dimensions of media content are evaluated, and the present disclosure obtains the value score of each dimension of media content respectively, and obtains the recommendation basis (content value score) of media content in combination with the value score of each dimension of media content, so as to recommend high-quality media content that meets the user's preferences to the user.

[0062] In one implementation of the present disclosure, Figure 3 As shown in the schematic diagram of the method for obtaining the content value score of media content in , the present disclosure adopts the following method to obtain the content value score of media content. Obtain media content and obtain the label of the media content. The obtained label of the media content can be directly used for the recommendation / operation of the media content. Among them, the content label includes factual labels, model labels and prediction labels. Factual labels include age, introduction, director, IP and producer. Model labels include type, subject matter, label weight, actor level and safety level. Prediction labels include initial rating of the appropriate audience, dynamic rating and content value score. Obtain the content quality value (first value score) according to factual labels (content labels). Obtain the platform traffic value (second value score) according to information such as holiday activity period, subject matter preference, willingness to pay, program exposure and income (traffic data). Determine the external emotional value (third value score) according to information such as the popularity of filmmakers and program popularity (heat data). Obtain the content value system according to the content quality value, platform traffic value and external emotional value, and then quantify the indicators to obtain the content value score. Among them, the content value score can be applied to the recommendation algorithm to observe the experimental effect, can be directly applied to the operation adjustment to observe the rationality of the adjustment, and can also predict the program payment conversion efficiency to observe the accuracy.

[0063] In one implementation of the disclosed embodiment, the value scores of different dimensions can be obtained through different models, such as Figure 4 As shown in the schematic diagram of the method for obtaining the content value score of media content, the data to be processed include: content quality (director, producer, initial rating, popular themes, ratings) (content tags), platform traffic (historical income, holidays, member activity cycles) (traffic data), public opinion heat (program heat, filmmaker heat, theater box office, want to watch index) (heat data). The models used to obtain the corresponding value score include linear regression model, SVM model, tree model and MLP model. For each dimension of data, after determining the corresponding model through feature selection, the content value score is obtained through model calculation.

[0064] The following embodiments of the present disclosure further illustrate data related to a media content.

[0065] In one implementation of the disclosed embodiment, the relevant data of the first media content includes one or more of the following data: the playback penetration rate of the media content, the payment conversion rate of the media content, or the user payment data when the media content is pushed a preset number of times.

[0066] In the disclosed embodiment, the playback penetration rate of media content is the proportion of the number of users browsing the content in the total number of users, which can represent the popularity of the media content. The payment conversion rate of media content reflects the user's willingness to pay for the media content, and can also represent the popularity of the media content. The user payment data of the media content when the preset number of pushes is used can represent the revenue of pushing the media content, and can also represent the popularity of the media content. Through the present disclosure, for the media content, the data that can reflect the popularity of the media content is used as the relevant data of the media content. Then, the recommendation basis of the media content is obtained based on the relevant data, so that the recommended media content meets the user's preferences.

[0067] Understandably, in daily life, there are often some unexpected events that gain huge popularity in a short period of time. However, the recommendation algorithm is often unable to identify these new contents and recommend them. The present disclosure can adjust the output of the algorithm through instructions to ensure that the media content that may gain huge popularity can be pushed to users in a timely manner.

[0068] Figure 5 FIG. 1 is a flow chart showing a method for setting first media content according to an exemplary embodiment. Figure 5 As shown, the method includes step S301 to step S302.

[0069] In step S301, first recommendation data is determined according to the content value score of the first media content and the relevant data of the first media content, where the first recommendation data includes the push priority of the first media content.

[0070] In step S302, based on the control instruction, the first preset media content is set and the priority is set for the first preset media content, the first preset media content is different from the first media content, and / or the recommendation data corresponding to the second preset media content is removed from the first recommendation data, and the first media content includes the second preset media content.

[0071] In the disclosed embodiment, the operator can adjust the first recommendation data through instructions, such as increasing or decreasing the push limit level of certain media content. It is also possible to directly set new media content (the same as the first preset media content of the first media content) in the recommendation data, set the priority for the new media content, or remove the recommendation data of some media content (the second preset media content) from the first recommendation content. In this way, the recommended content can be quickly adjusted according to specific situations such as current hot topics, festivals and celebrations, to ensure that media content that may gain huge popularity can be pushed to users in a timely manner, or to respond to emergencies and special dates / situations, to better reflect and convey specific values ​​and cultural orientations.

[0072] In one implementation of the present disclosure, Figure 6As shown in the schematic diagram of the method for setting recommended content according to control instructions, through the present disclosure, the operator can issue control instructions through the preset background window, adjust the weight through the algorithm, and set the manual intervention ratio. Through a specific window (channel today), determine the expected total UV resources and the expected remaining UV resources. The program pool can be manually intervened to select the program to be adjusted by serial number, program name or program ID, adjust the daily UV ratio, and determine the expected total exposure UV lower limit today (10,000), the expected remaining exposure UV lower limit today (10,000), today's exposure UV (10,000), today's broadcast UV (10,000) and today's sales (yuan) and other information.

[0073] It is understandable that the usage frequency of the video platform (smart network TV) is different in different time periods, and the traffic data generated by users browsing media content is also different. Therefore, the media content recommendation strength of the present disclosure is also different for different time periods.

[0074] Figure 7 FIG. 1 is a flow chart showing a method for pushing target media content according to an exemplary embodiment. Figure 7 As shown, the method includes steps S401 to S404.

[0075] In step S401, target flow data of a target time period is determined, where the target time period is the next time period adjacent to the current time period.

[0076] In step S402, the target recommendation times corresponding to the target traffic data are determined according to the corresponding relationship between the traffic data and the push times.

[0077] In step S403, target push data is determined according to current push data, a first number of currently online first category users, a second number of currently online second category users, a target recommendation number of times, and a first preset method.

[0078] Among them, the current push data includes a first push probability corresponding to a first type of user and a second push probability corresponding to a second type of user, and the target push data includes a first target push probability corresponding to the first type of user and a second push probability corresponding to the second type of user. The first type of user is a user who is interested in the target media content, and the second type of user is a user who is not interested in the media content.

[0079] In step S404, target media content is pushed to the first category of users and the second category of users according to the target push data.

[0080] In the disclosed embodiment, considering that the usage frequency of the video platform (intelligent network TV) in different time periods (such as working hours, off-get off work hours, holidays) is different, the traffic data generated by users browsing media content is also different. Therefore, the traffic data (target traffic data) of a future period (target time period) is predicted based on historical traffic data, and the corresponding number of recommendations (target recommendation number) is determined, and then content recommendations are made based on the predicted traffic data. Ensure that the traffic consumed by the recommended media content is consistent with the actual browsing situation of the user.

[0081] The following embodiments of the present disclosure illustrate a method for determining target push data.

[0082] Figure 8 FIG. 1 is a flow chart showing a method for determining target push data according to an exemplary embodiment. Figure 8 As shown, the method includes step S501, step S502A, step S502B and step S502C.

[0083] In step S501 , a first product of a first push probability and a first quantity is determined, and a second product of a second push probability and a second quantity is determined.

[0084] In step S502A, in response to the sum of the first product and the second product being equal to the target push number, the first push probability is determined as the first target push probability, and the second push probability is determined as the second target push probability.

[0085] In step S502B, in response to the sum of the first product and the second product being less than the target number of pushes, the first push probability is increased, and a third product of the increased first push probability and the first number is determined until the sum of the third product and the second product is equal to the target number of pushes, the increased first push probability is determined as the first target push probability, and the second push probability is determined as the second target push probability.

[0086] In step S502C, in response to the sum of the first product and the second product being greater than the target number of pushes, the second push probability is reduced, and a fourth product of the reduced second push probability and the second number is determined until the sum of the fourth product and the first product is equal to the target number of pushes, the reduced second push probability is determined as the second target push probability, and the first push probability is determined as the first target push probability.

[0087] In the disclosed embodiment, when the sum of the first product and the second product is equal to the target push number, the push number meets expectations, so the first push probability is determined as the first target push probability, and the second push probability is determined as the second target push probability.

[0088] In the disclosed embodiment, when the sum of the first product and the second product is less than the target push times, it indicates that the number of recommendations does not meet expectations, so the recommendation probability corresponding to the interested group is increased, and the target media content is recommended more times to the interested group. It can be understood that if the recommendation probability corresponding to the interested group (the first recommendation probability) rises to the maximum value (1), and the number of recommendations still does not meet expectations, the recommendation probability corresponding to the uninterested group (the second recommendation probability) is increased.

[0089] In the disclosed embodiment, when the sum of the first product and the second product is greater than the target push times, it indicates that the number of recommendations is too many, which will cause a waste of resources, so the recommendation probability corresponding to the uninterested group is reduced, and the target media content is recommended less times to the interested group. It can be understood that if the recommendation probability (second recommendation probability) corresponding to the uninterested group is reduced to the minimum value (0), and the number of recommendations is still higher than the target recommendation number, the recommendation probability (first recommendation probability) corresponding to the interested group is reduced.

[0090] In an exemplary embodiment of the present disclosure, the first target recommendation probability and the second target recommendation probability are set in the following manner:

[0091] (1) Users are divided into interested users and uninterested users, and the push probability of interested users in time period i is calculated. And push notifications to users who are not interested in time period i The expected media content delivery is as follows:

[0092]

[0093] Among them, i represents a certain time period, is the number of users interested in watching media content in time period i (the first number of users of the first category), is the number of users who are not interested in watching media content in time period i (the second number of second category users).

[0094] (2) Assuming that the proportion of users interested in watching media content in the next time period i+1 is the same as that in the previous time period, the estimated number of users interested in watching media content in time period i+1 is:

[0095]

[0096] Where i+1 is the next time period corresponding to time period i, The estimated number of users interested in watching media content for time period i+1, num i is the total number of users in time period i, and num i+1 is the total number of users in interval i+1.

[0097] Similarly, the expected mean number of exposed users in time period i+1 is:

[0098]

[0099] in, is the expected mean number of exposed users in time period i+1, is the push probability for users who are interested in watching media content in time period i+1, is the push probability for users who are not interested in watching media content in time period i+1, is the number of users interested in watching media content in time period i+1, num i+1 The total number of users in time period i+1.

[0100] (3) Calculation results:

[0101]

[0102] Among them, expected_uv is the expected exposure data, exposed_uv is the exposed data, num i+1 is the total number of users in time period i+1, num j is the total number of users in the time period after time period i+1, T is the time period after time period i+1, then is the total number of users corresponding to time period i+1 to time period T, is the push probability for users who are interested in watching media content in time period i+1, is the push probability for users who are not interested in watching media content in time period i+1, Estimate the number of users interested in watching media content for time period i+1.

[0103] Further calculation yields:

[0104]

[0105] Among them, expected_uv is the expected exposure data, exposed_uv is the exposed data, num j is the total number of users in the time period after time period i+1, T is the time period after time period i+1, then is the total number of users corresponding to time period i+1 to time period T, is the push probability for users who are interested in watching media content in time period i+1, is the push probability for users who are not interested in watching media content in time period i+1, num i is the total number of users in time period i, is the number of users interested in watching media content in time period i.

[0106] The above formula is simplified to:

[0107]

[0108] Among them, a is the number of users who are interested in watching media content, b is the number of users who are not interested in watching media content, is the push probability for users who are interested in watching media content in time period i+1, is the push probability for users who are not interested in watching media content in time period i+1, and c is the total number of users.

[0109] (4) Finally, it is transformed into an online learning problem. Based on the data from the previous time period, the and There are three situations:

[0110] Keep the probability of sending to interested users and uninterested users unchanged

[0111] Prioritize increasing the probability of interested users (first push probability), and then raise the probability of uninterested users when it exceeds 1

[0112] Prioritize reducing the probability of uninterested users (second push probability), and then reduce the probability of interested users when it is less than 0.

[0113] The following embodiments of the present disclosure illustrate a method for determining target flow data for a target time period.

[0114] Fig. 9 FIG. 1 is a flow chart showing a method for determining target flow data for a target time period according to an exemplary embodiment. Fig. 9 As shown, the method includes steps S601 to S603.

[0115] In step S601, data fitting is performed for a target time period based on the first historical traffic data through a preset model to obtain fitting traffic data corresponding to the target time period. The first historical traffic data includes traffic data corresponding to different time periods within a first cycle from the target time period.

[0116] In step S602, the second historical traffic data of the historical time period is determined, and the historical time period is fitted with data through a preset model to obtain the historical fitting traffic data corresponding to the historical time period. The historical time period is the time period between 0:00 on the day and the preset time interval before the current time period.

[0117] In step S603, the fitting flow data is adjusted according to the difference ratio between the second historical flow data and the historical fitting flow data to obtain the target flow data.

[0118] In the disclosed embodiment, the time series fitting adopts a preset model, and eliminates the true value error by rolling calculation of the data in the first cycle. The historical flow data for the historical time period (such as 0 o'clock on the day to 2 hours before the current time point) and the time series flow data for the same period are used for calculation to obtain the difference ratio between the fitting data and the true data. The residual flow obtained by the time series fitting is multiplied by the coefficient to perform the residual flow adjustment calculation, and the adjusted residual flow (target flow data) can be obtained. Through the present disclosure, the overall flow prediction result obtained is very close to the true value, which well solves the problem of inaccurate flow prediction on specific dates (such as holidays) that may occur.

[0119] In one implementation of the present disclosure, Fig.10 As shown in the schematic diagram of the method for determining the target traffic data of the target time period, the target traffic data is determined in the following way: the overall mechanism adopts the scheme of time series fitting + mean value as a bottom line, and the time series fitting adopts the Prophet model (preset model). By rolling calculation of the data of the historical year (first cycle), the error of the true value is eliminated; and on this basis, further coefficient tuning is performed, and the true value (historical traffic data) from 0 o'clock on the day to 2 hours before the current time point (considering delayed reporting) and the time series fitting value (fitted traffic data) of the same period are used for calculation to obtain a difference ratio, which is the difference ratio between the fitting data and the true data. The remaining traffic obtained by time series fitting is multiplied by the coefficient for residual traffic adjustment calculation, and the adjusted remaining traffic (target durian data) can be obtained. In this way, the new coefficient and the corresponding remaining traffic are calculated once every hour, and the overall traffic forecast result obtained is very close to the true value, which solves the problem of inaccurate holiday traffic forecast that may occur. Considering that time series forecasting has high requirements for the real-time output of data, the overall accuracy of the engine will be affected if the forecast data is not produced in time, so the mean forecast is adopted as a bottom line solution. The mean prediction scheme estimates the remaining traffic based on the mean of weekdays or weekends in the past week, and reduces the true value error caused by historical data as much as possible by shortening the data recovery cycle. The actual prediction results are basically close to the true value.

[0120] In an exemplary embodiment of the present disclosure, Fig.11 As shown in the schematic diagram of historical traffic data, the historical traffic data includes traffic data corresponding to different time periods in a past period of time.

[0121] In an exemplary embodiment of the present disclosure, Fig.12As shown in the schematic diagram of the fitted data and the actual data, there will be a deviation between the flow data fitted by the preset model and the actually collected flow data. Therefore, the fitted flow data should be corrected according to the actually collected flow data to make it closer to the actual situation.

[0122] In an exemplary embodiment of the present disclosure, Fig.13 As shown in the schematic diagram of the media content recommendation method, the scheme includes the following steps: obtaining media resources of TV series, movies, animation, variety shows, children, vertical categories, etc., and using media resources and real-time monitoring data (play penetration rate, payment conversion rate, and payment per thousand exposures) for recommendation. Determine the content value system based on the media resources combined with ratings, quality, labels, playback UV and payment data, and provide recommendation basis. Manual intervention is directly implemented on the algorithm based on the content and recommendation basis output by the recommendation module, and the algorithm output is empowered (three-stage rocket, content planning, old drama re-promotion, user grouping, public opinion trend, material rotation) to obtain output data. The algorithm is processed based on the content and recommendation basis output by the recommendation module and the real-time intervention of manual intervention (model iteration, portrait system, content recall, user identity), and is manually empowered to obtain output. Traffic is allocated for data, content is recommended for user terminals, and data is monitored in real time for user terminals. Among them, the media content recommendation method includes six core contents: 1. Manual intervention configuration: the configuration background provides the channel content manual intervention configuration page; 2. Content value score evaluation: build value score labels and apply them to manual intervention and algorithm recommendation scenarios; 3. Intervention traffic prediction: real-time remaining traffic estimation to maximize traffic utilization efficiency; 4. Intelligent traffic distribution: evenly distribute guaranteed programs according to the manual intervention traffic ratio; 5. Scientific experimental verification: verify the effects of different recommendation modules such as manual intervention and pure algorithm recommendation; 6. Visual real-time data dashboard: the configuration background and data platform provide real-time feedback dashboard to ensure that manual intervention is fast, correct and accurate. Through this solution, a new platform model that cooperates with the algorithm is established from 0 to 1 to achieve the system goal of efficient, safe and scientific intervention. To achieve key capability construction, in terms of system efficiency improvement: automatic capabilities such as timed stop of intervention and timed release of fixed positions; daily report of intervention system data based on channels. In terms of configuration security: fool-proof strategy for blocking configuration in situations such as abnormal content status and lack of necessary materials; the ability to remind the responsible person in abnormal situations such as content offline; the ability to quickly send data to internal devices for preview after the intervention is completed. In terms of scientific human intervention: content selection scenarios; content value scores, sales forecasts and other data displays help humans judge the value level of content; the system actively captures recent popular programs, themes, tags and other dimensional fields to provide reference; intervention configuration scenarios; real-time traffic estimates while providing the best intervention ratio (ECPM highest) prediction to help humans better set intervention rules.

[0123] In an exemplary embodiment of the present disclosure, Fig.14 1 is a schematic diagram of a media content recommendation architecture according to an exemplary embodiment of the present disclosure. The overall architecture of the media content recommendation system in this solution includes an application layer, a gateway layer, an online service layer, an operation layer, and a data processing layer. The contents corresponding to the gray boxes in the online service layer, the operation layer, and the data processing layer in the figure are the service contents involved in the media recommendation method in the present disclosure.

[0124] In an embodiment of the present disclosure, for a first media content, a content value score representing a first quality of the media content is obtained, and related data representing the popularity of the media content is obtained. Based on the content value score and related data of the first media content, first recommendation data is determined, and the first recommendation data includes a push priority of the first media content. Based on the received control instruction, the push priority of the first media content is adjusted, and second recommendation data is determined, and the second recommendation data includes a push priority of second media content, and the second media content is different from the first media content and includes multiple media contents. Based on the push priority of the second media content and a preset priority threshold, the target media content is determined and pushed. Through the present disclosure, while ensuring that content recommendation is performed efficiently through an algorithm and meets the personalized requirements of users, new hot content can be recommended through manual intervention.

[0125] In the disclosed embodiment, the entire link of the entire engine is opened up, the configuration platform, content value and server-side architecture are successfully upgraded, and the "quantity guarantee model" link is introduced on the previous algorithm recommendation architecture and put into use.

[0126] In the disclosed embodiment, an innovative intelligent traffic engine for Internet TV is proposed, which combines algorithm recommendation with manual weight adjustment, and realizes the maximum rational distribution of traffic through the strategy of manual assistance algorithm, improves the efficiency of content distribution, and thus drives the improvement of playback penetration and the increase of payment per thousand exposures. It aims to realize a new algorithm recommendation model for Internet TV, effectively integrate the manual and algorithm recommendation models, and give full play to their respective advantages in the content recommendation process, realize the integration of people and algorithms, and achieve the optimal solution for the content recommendation model. The engine designed in the present disclosure is mainly used in the content recommendation scenario of smart Internet TV: after the user turns on the device and browses the content data between channels, the engine will combine the manual intervention configuration of the current channel and the recommendation exposure of the algorithm, and after complex calculations and traffic allocation, make the final personalized content recommendation for the user.

[0127] Based on the same concept, the embodiment of the present disclosure also provides a media content recommendation device 100 .

[0128] It is understandable that the media content recommendation device 100 provided in the embodiment of the present disclosure includes hardware structures and / or software modules corresponding to the execution of each function in order to realize the above functions. In combination with the units and algorithm steps of each example disclosed in the embodiment of the present disclosure, the embodiment of the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the technical solution of the embodiment of the present disclosure.

[0129] Fig.15 is a block diagram of a media content recommendation device 100 according to an exemplary embodiment. Fig.15 The device includes an acquisition unit 101, a determination unit 102, a processing unit 103 and a pushing unit 104.

[0130] The acquisition unit 101 is used to acquire the content value score of the first media content and acquire relevant data of the first media content. The first media content includes multiple media contents. The content value score represents the comprehensive quality of the media content, and the relevant data represents the popularity of the media content.

[0131] The determination unit 102 is configured to determine first recommendation data according to the content value score of the first media content and relevant data of the first media content, where the first recommendation data includes a push priority of the first media content.

[0132] The processing unit 103 is used to determine the second recommendation data according to the first recommendation data and the received control instruction, the second recommendation data including the push priority of the second media content, the control instruction is used to adjust the push priority of the first media content, and the second media content is different from the first media content and includes multiple media content.

[0133] The pushing unit 104 is configured to determine and push target media content according to the push priority of the second media content and a preset priority threshold.

[0134] In one implementation, the acquisition unit 101 acquires the content value score of the first media content in the following manner: for each of the multiple media contents in the first media content, determine the content tag, traffic data, and popularity data of the media content. Determine the first value score of the media content based on the content tag, the first value score represents the content quality of the media content, determine the second value score of the media content based on the traffic data, the second value score represents the platform traffic occupied by the media content, and determine the third value score of the media content based on the popularity data, the second value score represents the online popularity of the media content. The sum of the first value score, the second value score, and the third value score is determined as the content value score of the media content.

[0135] In one implementation, the data related to the first media content includes one or more of the following data: a playback penetration rate of the media content, a payment conversion rate of the media content, or user payment data when the media content is pushed a preset number of times.

[0136] In one implementation, the processing unit 103 is further configured to: based on the control instruction, set the first preset media content and set a priority for the first preset media content, the first preset media content being different from the first media content. And / or remove the recommendation data corresponding to the second preset media content from the first recommendation data, the first media content including the second preset media content.

[0137] In one embodiment, the push unit 104 pushes the target media content in the following manner: determine the target traffic data of the target time period, where the target time period is the next time period adjacent to the current time period. Determine the target recommendation number corresponding to the target traffic data based on the correspondence between the traffic data and the number of pushes. Determine the target push data based on the current push data, the first number of the first category of users currently online, the second number of the second category of users currently online, the target recommendation number, and the first preset method, wherein the current push data includes a first push probability corresponding to the first category of users and a second category of push probability corresponding to the second category of users, and the target push data includes a first target push probability corresponding to the first category of users and a second category of push probability corresponding to the second category of users, the first category of users are users who are interested in the target media content, and the second category of users are users who are not interested in the media content. According to the target push data, push the target media content to the first category of users and the second category of users.

[0138] In one embodiment, the push unit 104 determines the target push data according to the current push data, the first number of the first category of users currently online, the second number of the second category of users currently online, the target push number and the first preset method in the following manner: determine the first product of the first push probability and the first number, and determine the second product of the second push probability and the second number. In response to the sum of the first product and the second product being equal to the target push number, determine the first push probability as the first target push probability, and determine the second push probability as the second target push probability. In response to the sum of the first product and the second product being less than the target push number, increase the first push probability, and determine the third product of the increased first push probability and the first number, until the sum of the third product and the second product is equal to the target push number, determine the increased first push probability as the first target push probability, and determine the second push probability as the second target push probability. In response to the sum of the first product and the second product being greater than the target number of pushes, the second push probability is reduced, and a fourth product of the reduced second push probability and the second number is determined until the sum of the fourth product and the first product is equal to the target number of pushes, the reduced second push probability is determined as the second target push probability, and the first push probability is determined as the first target push probability.

[0139] In one embodiment, the push unit 104 determines the target flow data of the target time period in the following manner: through a preset model, based on the first historical flow data, data fitting is performed for the target time period to obtain fitting flow data corresponding to the target time period, and the first historical flow data includes flow data corresponding to different time periods within a first cycle from the target time period. The second historical flow data of the historical time period is determined, and data fitting is performed for the historical time period through the preset model to obtain historical fitting flow data corresponding to the historical time period, and the historical time period is the time period between 0:00 on the day and the preset time interval before the current time period. According to the difference ratio between the second historical flow data and the historical fitting flow data, the fitting flow data is adjusted to obtain the target flow data.

[0140] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0141] Fig.16 1 is a block diagram of a device 200 for media content recommendation according to an exemplary embodiment. The device 200 may be provided as a terminal. For example, the device 200 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0142] Reference Fig.16, the device 200 may include one or more of the following components: a processing component 202 , a memory 204 , a power component 206 , a multimedia component 208 , an audio component 210 , an input / output (I / O) interface 212 , a sensor component 214 , and a communication component 216 .

[0143] The processing component 202 generally controls the overall operation of the device 200, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 202 may include one or more processors 220 to execute instructions to perform all or part of the steps of the above-described method. In addition, the processing component 202 may include one or more modules to facilitate interaction between the processing component 202 and other components. For example, the processing component 202 may include a multimedia module to facilitate interaction between the multimedia component 208 and the processing component 202.

[0144] The memory 204 is configured to store various types of data to support operations on the device 200. Examples of such data include instructions for any application or method operating on the device 200, contact data, phone book data, messages, pictures, videos, etc. The memory 204 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0145] The power component 206 provides power to the various components of the device 200. The power component 206 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 200.

[0146] The multimedia component 208 includes a screen that provides an output interface between the device 200 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 208 includes a front camera and / or a rear camera. When the device 200 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0147] The audio component 210 is configured to output and / or input audio signals. For example, the audio component 210 includes a microphone (MIC), and when the device 200 is in an operation mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 204 or sent via the communication component 216. In some embodiments, the audio component 210 also includes a speaker for outputting audio signals.

[0148] I / O interface 212 provides an interface between processing component 202 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0149] The sensor assembly 214 includes one or more sensors for providing various aspects of the status assessment of the device 200. For example, the sensor assembly 214 can detect the open / closed state of the device 200, the relative positioning of components, such as the display and keypad of the device 200, the sensor assembly 214 can also detect the position change of the device 200 or a component of the device 200, the presence or absence of user contact with the device 200, the orientation or acceleration / deceleration of the device 200 and the temperature change of the device 200. The sensor assembly 214 can include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 214 can also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 214 can also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor or a temperature sensor.

[0150] The communication component 216 is configured to facilitate wired or wireless communication between the device 200 and other devices. The device 200 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 216 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 216 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0151] In an exemplary embodiment, the apparatus 200 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0152] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 204 including instructions, and the instructions can be executed by the processor 220 of the device 200 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0153] Fig.17 is a block diagram of a device 300 for media content recommendation according to an exemplary embodiment. For example, the device 300 may be provided as a server. Fig.17 The apparatus 300 includes a processing component 322, which further includes one or more processors, and a memory resource represented by a memory 332 for storing instructions, such as an application, that can be executed by the processing component 322. The application stored in the memory 332 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 322 is configured to execute instructions to perform the above-mentioned media content recommendation method.

[0154] The device 300 may also include a power supply component 326 configured to perform power management of the device 300, a wired or wireless network interface 350 configured to connect the device 300 to a network, and an input / output (I / O) interface 358. The device 300 may operate based on an operating system stored in the memory 332, such as Windows Server TM, Mac OS X TM, Unix TM, Linux TM, FreeBSD TM or the like.

[0155] It is to be understood that in the present disclosure, "plurality" refers to two or more than two, and other quantifiers are similar. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. The singular forms "a", "the" and "the" are also intended to include plural forms, unless the context clearly indicates other meanings.

[0156] It is further understood that the terms "first", "second", etc. are used to describe various information, but such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other, and do not indicate a specific order or degree of importance. In fact, the expressions "first", "second", etc. can be used interchangeably. For example, without departing from the scope of the present disclosure, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information.

[0157] It will be further understood that the terms “center”, “longitudinal”, “lateral”, “front”, “back”, “up”, “down”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the present embodiment and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation.

[0158] It can be further understood that, unless otherwise specified, “connection” includes a direct connection without other components between the two, and also includes an indirect connection with other components between the two.

[0159] It is further understood that, although the operations are described in a specific order in the drawings in the embodiments of the present disclosure, it should not be understood as requiring the operations to be performed in the specific order shown or in a serial order, or requiring the execution of all the operations shown to obtain the desired results. In certain environments, multitasking and parallel processing may be advantageous.

[0160] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present solution, which follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure.

[0161] It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the scope of the appended claims.

Claims

1. A media content recommendation method, characterized in that: include: Obtaining a content value score of a first media content and obtaining relevant data of the first media content, wherein the first media content includes multiple media contents, the content value score represents the comprehensive quality of the media content, and the relevant data represents the popularity of the media content; determining first recommendation data according to the content value score of the first media content and the relevant data of the first media content, wherein the first recommendation data includes a push priority of the first media content; determining second recommendation data according to the first recommendation data and the received control instruction, wherein the second recommendation data includes a push priority of second media content, the control instruction is used to adjust the push priority of the first media content, and the second media content is different from the first media content and includes multiple media content; According to the push priority of the second media content and a preset priority threshold, target media content is determined and pushed.

2. The method according to claim 1, characterized in that The obtaining of the content value score of the first media content includes: For each media content in the plurality of media content in the first media content, determine a content tag, flow data, and popularity data of the media content; Determine a first value score of the media content according to the content tag, the first value score represents the content quality of the media content, determine a second value score of the media content according to the traffic data, the second value score represents the platform traffic occupied by the media content, and determine a third value score of the media content according to the popularity data, the second value score represents the online popularity of the media content; The sum of the first value score, the second value score and the third value score is determined as the content value score of the media content.

3. The method according to claim 1, characterized in that The data related to the first media content includes one or more of the following data: the playback penetration rate of the media content, the payment conversion rate of the media content, or the user payment data when the media content is pushed a preset number of times.

4. The method according to claim 1, characterized in that: The method further comprises: Based on the control instruction, setting a first preset media content and setting a priority for the first preset media content, the first preset media content being different from the first media content; and / or The recommended data corresponding to the second preset media content is removed from the first recommended data, and the first media content includes the second preset media content.

5. The method according to claim 1, characterized in that: The push target media content includes: Determine target flow data for a target time period, where the target time period is a next time period adjacent to the current time period; Determine the target recommendation number corresponding to the target traffic data according to the corresponding relationship between the traffic data and the push number; Determine the target push data according to the current push data, the first number of the first category of users currently online, the second number of the second category of users currently online, the target recommendation times and the first preset method, wherein the current push data includes a first push probability corresponding to the first category of users and a second push probability corresponding to the second category of users, the target push data includes a first target push probability corresponding to the first category of users and a second push probability corresponding to the second category of users, the first category of users are users who are interested in the target media content, and the second category of users are users who are not interested in the media content; The target media content is pushed to the first category of users and the second category of users according to the target push data.

6. The method according to claim 5, characterized in that The determining the target push data according to the current push data, the first number of the first category of users currently online, the second number of the second category of users currently online, the target push times and the first preset method includes: Determine a first product of a first push probability and the first number, and determine a second product of a second push probability and the second number; In response to the sum of the first product and the second product being equal to the target number of pushes, determining the first push probability as a first target push probability, and determining the second push probability as a second target push probability; In response to the sum of the first product and the second product being less than the target number of pushes, the first push probability is increased, and a third product of the increased first push probability and the first number is determined, until the sum of the third product and the second product is equal to the target number of pushes, the increased first push probability is determined as the first target push probability, and the second push probability is determined as the second target push probability; In response to the sum of the first product and the second product being greater than the target push number, the second push probability is reduced, and a fourth product of the reduced second push probability and the second number is determined, until the sum of the fourth product and the first product is equal to the target push number, the reduced second push probability is determined as the second target push probability, and the first push probability is determined as the first target push probability.

7. The method according to claim 5, characterized in that The target flow data for the target time period is determined, including: By using a preset model, data fitting is performed for the target time period based on the first historical flow data to obtain fitting flow data corresponding to the target time period, wherein the first historical flow data includes flow data corresponding to different time periods within a first cycle from the target time period; Determine the second historical traffic data of the historical time period, and perform data fitting on the historical time period through a preset model to obtain the historical fitting traffic data corresponding to the historical time period, wherein the historical time period is the time period between 0:00 on the day and a preset time interval before the current time period; According to the difference ratio between the second historical flow data and the historical matching flow data, the matching flow data is adjusted to obtain the target flow data.

8. A media content recommendation device, characterized in that: include: an acquisition unit, configured to acquire a content value score of a first media content and acquire relevant data of the first media content, wherein the first media content includes a plurality of media contents, the content value score represents the comprehensive quality of the media content, and the relevant data represents the popularity of the media content; a determining unit, configured to determine first recommendation data according to the content value score of the first media content and the relevant data of the first media content, wherein the first recommendation data includes a push priority of the first media content; a processing unit, configured to determine second recommendation data according to the first recommendation data and the received control instruction, wherein the second recommendation data includes a push priority of second media content, the control instruction is used to adjust the push priority of the first media content, and the second media content is different from the first media content and includes multiple media content; The pushing unit is used to determine and push the target media content according to the pushing priority of the second media content and a preset priority threshold.

9. An electronic device, characterized in that: include: processor: a memory for storing processor-executable instructions; The processor is configured to: execute the media content recommendation method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores instructions, and when the instructions in the storage medium are executed by a processor, the processor is enabled to execute the media content recommendation method according to any one of claims 1 to 7.

Images