Intelligent data analysis and processing method for film and television big data platform
By analyzing the interactive data of users on the film and television big data platform, calculating the degree of preference and correction of video types, the problem that recommendation algorithms in the prior art cannot meet users' personalized needs is solved, and more accurate video recommendations are achieved.
Patent Information
- Application Number
- CN202410983308.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-08
AI Technical Summary
The existing film and television data analysis and processing mode cannot meet the personalized needs of users. The simple and crude recommendation algorithm only relies on the popularity dimension and cannot provide more accurate content recommendations.
By obtaining the total length of the video, the length of viewing of the user in different time periods, and the interactive data (likes, comments, forwardings, barrages, and coin count), the user's interactive expression, likes, and corrections for the video type. Then determine the video type of users' final preference and optimize the recommendation algorithm.
It provides content recommendations that are more in line with user interests and behavior patterns, realizes personalized video recommendations, and improves user satisfaction and viewing experience.
Smart Images

Figure CN118921510A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and in particular to a method for intelligent analysis and processing of data on a film and television big data platform. Background Art
[0002] The film and television big data platform is a comprehensive platform designed to meet the data management, analysis and application needs of the film and television industry. It can integrate and process large-scale data from various data sources, and provide valuable insights and services through data analysis and mining technology to support decision-making and operations in the production, promotion, distribution and marketing of film and television content. With the comprehensive development of technology, science, economy and culture, people have begun to pay attention to the pursuit of spiritual and cultural time, and film and television entertainment has assumed an important role as a carrier and has begun to attract widespread attention. The existing film and television data analysis and processing model can no longer meet people's needs.
[0003] The data intelligent analysis and processing of existing film and television platforms is a relatively crude film and television recommendation algorithm. Based solely on the data on the popularity dimension of the drama, it is believed that everyone tends to chase popular TV dramas all the time, occupying the video homepage for a long time, affecting users' viewing of other film and television resources. This simple and crude recommendation algorithm can no longer meet the needs of users. Summary of the invention
[0004] The present invention provides a method for intelligent analysis and processing of film and television big data platform data to solve the existing problems.
[0005] The present invention provides a method for intelligent analysis and processing of film and television big data platform data using the following technical solutions:
[0006] An embodiment of the present invention provides a method for intelligent analysis and processing of data on a film and television big data platform, the method comprising the following steps:
[0007] Get the total duration of each video within the preset total number of days; divide each day into several time periods; get the viewing time of each video of each video type watched by the user in each time period of each day, the number of all videos of each video type watched by the user in each time period of each day, and the number of likes, comments, reposts, bullet comments, and coins for each video type by the user in each time period of each day;
[0008] According to the number of likes, comments, reposts, bullet comments, and coins for each video type in each time period of the day, the user's interactive expression for each video type in each time period of the day is obtained; according to the user's interactive expression for each video type in each time period of the day, the viewing time and total viewing time of all videos of each video type, the user's liking degree for each video type in each time period of the day is obtained;
[0009] Based on the user's preference for each video type in each time period of all days and the time the user spends watching videos of all video types in each time period of each day, the user's correction degree for each video type in each time period is obtained; based on the user's correction degree for each video type in each time period and the user's preference for each video type in each time period of all days, the user's final preference for each video type in each time period is obtained; based on the user's final preference for each video type in each time period, the user's favorite video type in each time period is obtained.
[0010] Furthermore, according to the number of likes, comments, reposts, bullet screens, and coins of each video type by the user in each time period of each day, a specific calculation formula for the user's interactive expressiveness for each video type in each time period of each day is obtained as follows:
[0011] b v,k,x =y1a v,k,x1 +y2a v,k,x2 +y3a ,k,x3 +y4a v,k,x4 +y5a v,k,x5
[0012] Where b v,k,x represents the user's interactive expression on the xth video type in the kth time period on the vth day; y1 represents the preset like coefficient; a v,k,x1 represents the number of likes given by the user to the xth video type in the kth time period on the vth day; y2 represents the preset comment coefficient; a v,k,x2 represents the number of comments made by the user on the xth video type in the kth time period on the vth day; y3 represents the preset forwarding coefficient; a v,k,x3 represents the number of times the user forwarded the xth video type in the kth time period on the vth day; y4 represents the preset bullet screen coefficient; a v,k,x4 represents the number of comments made by the user on the xth video type in the kth time period on the vth day; y5 represents the preset coin coefficient; a v,k,x5 It indicates the number of times the user pays coins to the xth video type in the kth time period on the vth day.
[0013] Furthermore, the method of obtaining the user's preference for each video type in each time period of the day according to the user's interactive expression for each video type in each time period of the day, the viewing time and the total viewing time of all videos of each video type, includes the following specific steps:
[0014] According to the viewing time and total viewing time of all videos of each video type watched by users in each time period of each day, the user's attention to each video type in each time period of each day is obtained;
[0015] According to the user's interactive expression and attention to each video type in each time period of the day, the user's preference degree for each video type in each time period of the day is obtained.
[0016] Furthermore, according to the viewing time and total viewing time of all videos of each video type watched by the user in each time period of each day, a specific calculation formula for obtaining the user's attention to each video type in each time period of each day is:
[0017]
[0018] In the formula, c v,k,x represents the user's attention to the xth video type in the kth time period on the vth day; I v,k,x represents the number of videos of the xth video type watched by the user in the kth time period on the vth day; m v,k,x,i M represents the viewing time of the i-th video of the x-th video type in the k-th time period of the v-th day; v,k,x,i It indicates the total time that the user watches the i-th video of the x-th video type in the k-th time period on the v-th day.
[0019] Furthermore, the method of obtaining the user's preference for each video type in each time period of each day according to the user's interactive expression and attention to each video type in each time period of each day includes the following specific steps:
[0020] The product of the user's interactive expression and attention to each video type in each time period of the day is recorded as the user's liking for each video type in each time period of the day.
[0021] Furthermore, the method of obtaining the user's correction degree for each video type in each time period according to the user's liking degree for each video type in each time period of all days and the time the user watches videos of all video types in each time period of each day includes the following specific steps:
[0022] Within the preset total number of days, with the time period as the horizontal axis and the user's preference for the x-th video type in each time period as the vertical axis, a curve graph corresponding to the x-th video type is constructed;
[0023] According to the user's preference for each video type in each time period of all days, the time the user watches videos of all video types in each time period of each day, and the curve graph corresponding to each video type, the user's correction degree for each video type in each time period is obtained.
[0024] Furthermore, according to the user's preference for each video type in each time period of all days, the time the user watches videos of all video types in each time period of each day, and the curve graph corresponding to each video type, a specific calculation formula for obtaining the user's correction degree for each video type in each time period is:
[0025]
[0026] In the formula, e k,x represents the degree of correction of the xth video type by the user in the kth time period; X represents the total number of preset days; d v,k,x N represents the user's preference for the xth video type in the kth time period on the vth day; v,k represents the sum of the time that the user watches all videos of all video types in the kth time period on the vth day; n v,k,x represents the sum of the time that the user watches all videos of the xth video type in the kth time period on the vth day; k v,x It represents the slope of the data point corresponding to the user in the kth time period on the vth day on the curve graph corresponding to the xth video type; sigmoid() is a linear normalization function.
[0027] Furthermore, the method of obtaining the final liking degree of the user for each video type in each time period according to the correction degree of the user for each video type in each time period and the liking degree of the user for each video type in each time period of all days includes the following specific steps:
[0028] The average of the user's liking for each video type in each time period of all days is recorded as the user's liking for each video type in each time period;
[0029] According to the correction degree and the liking degree of the user for each video type in each time period, the final liking degree of the user for each video type in each time period is obtained.
[0030] Furthermore, the specific calculation formula for obtaining the user's final liking degree for each video type in each time period according to the user's correction degree and liking degree for each video type in each time period is:
[0031]
[0032] In the formula, r k,x represents the user's final preference for the xth video type in the kth time period; e k,x represents the degree of correction of the xth video type by the user in the kth time period; d k,x represents the user's preference for the xth video type in the kth time period; O represents the number of time periods in a day; do,x Indicates the user's preference for the xth video type in the oth time period.
[0033] Furthermore, the method of obtaining the user's favorite video type in each time period according to the user's final liking degree for each video type in each time period includes the following specific steps:
[0034] The video type corresponding to the maximum value of the user's final liking degree for each video type in each time period is recorded as the user's favorite video type in each time period.
[0035] The beneficial effects of the technical solution of the present invention are:
[0036] In an embodiment of the present invention, within a preset total number of days, the total duration of each video is obtained; each day is equally divided into several time periods; the viewing time of each video of each video type watched by the user in each time period of each day, the number of all videos of each video type watched by the user in each time period of each day, and the number of likes, comments, reposts, barrages, and coins for each video type by the user in each time period of each day are obtained, so as to provide a data basis for subsequent data analysis; based on the number of likes, comments, reposts, barrages, and coins for each video type by the user in each time period of each day, the user's interactive expressiveness for each video type in each time period of each day is obtained, and the user's degree of interaction and participation in each video type is evaluated; based on the user's interactive expressiveness for each video type in each time period of each day, the viewing time and total time of all videos of each video type watched by the user, the user's The degree of liking of each video type helps to adjust the recommendation algorithm and provide content recommendations that are more in line with the user's interests; according to the user's degree of liking each video type in each time period of all days and the time the user watches videos of all video types in each time period of every day, the user's correction degree for each video type in each time period is obtained, which is convenient for correcting the user's degree of liking each video type in each time period of every day; according to the user's correction degree for each video type in each time period and the user's degree of liking each video type in each time period of all days, the user's final degree of liking for each video type in each time period is obtained, which is convenient for determining the user's final favorite video type in each time period; according to the user's final degree of liking each video type in each time period, the user's favorite video type in each time period is obtained, which is used for the optimization of the real-time recommendation system to ensure that the content recommendation is more in line with the user's current interests and behavior patterns. So far, the present invention analyzes the user's preferences for watching videos in history, customizes personalized algorithm recommendation video types for users, and makes more accurate video recommendations. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0038] Figure 1 The present invention is a flowchart of the steps of a method for intelligent analysis and processing of data on a film and television big data platform. DETAILED DESCRIPTION
[0039] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, characteristics and effects of a film and television big data platform data intelligent analysis and processing method proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0041] The following is a detailed description of a method for intelligent analysis and processing of film and television big data platform data provided by the present invention in conjunction with the accompanying drawings.
[0042] See also Figure 1 , which shows a flowchart of a method for intelligent analysis and processing of data on a film and television big data platform provided by an embodiment of the present invention, the method comprising the following steps:
[0043] Step S001: Get the total duration of each video within the preset total number of days; divide each day into several time periods; get the viewing time of each video of each video type watched by the user in each time period of the day, the number of all videos of each video type watched by the user in each time period of the day, and the number of likes, comments, reposts, barrages, and coins for each video type by the user in each time period of the day.
[0044] In the database of a video website platform, taking any user as an example, obtain the total duration of each video within a preset total number of days;
[0045] Divide each day into several time periods; obtain the viewing time of each video of each video type watched by the user in each time period of the day, the number of all videos of each video type watched by the user in each time period of the day, and the number of likes, comments, reposts, barrages, and coins for each video type by the user in each time period of the day.
[0046] It should be noted that the preset total time period in this embodiment is ten days, and each day is divided into three time periods, which is used as an example for description; within the preset total number of days, the user may watch the same video repeatedly or in different time periods, so the sum of all the time the user watches a video is the viewing time of each video; by analyzing the viewing data of users within the preset total time period, their viewing habits and preferred video types can be understood. Users prefer different video types in different time periods. For example, they prefer news videos in the morning, food videos at lunch, and animation videos in the evening. This personalized analysis can optimize the video recommendation algorithm, ensure that the recommendation is more in line with the user's preferences, and improve user satisfaction and viewing experience.
[0047] Step S002: According to the number of likes, comments, reposts, barrages, and coins of each video type by the user in each time period of the day, the user's interactive expressiveness for each video type in each time period of the day is obtained; according to the user's interactive expressiveness for each video type in each time period of the day, the viewing time and total viewing time of all videos of each video type, the user's liking degree for each video type in each time period of the day is obtained.
[0048] The same user has different preferences for videos of different styles and types. The user's preference for videos can be reflected through the user's video interaction, etc. Different video interaction methods represent different user psychology. According to the different silent costs required for different interaction methods, video interaction is divided into likes, comments, reposts, bullet screens, and coins. Different interaction methods represent different user psychology. According to the different silent costs (time, coins) required for different interaction methods, the weight of the coin interaction behavior is greater than that of other interaction behaviors.
[0049] In this embodiment, the preset like coefficient y1=0.1, comment coefficient y2=0.1, forwarding coefficient y3=0.1, bullet screen coefficient y4=0.2, and coin coefficient y5=0.5 are described as an example. Other implementation methods may be set to other values, which are not limited in this embodiment;
[0050] According to the number of likes, comments, reposts, bullet screens, and coins for each video type in each time period of the day, the corresponding calculation formula for the user's interactive expression for each video type in each time period of the day is:
[0051] b v,k,x =y1a v,k,x1 +y2a v,k,x2 +y3a v,k,x3 +y4a v,k,x4 +y5a v,k,x5
[0052] Where b v,k,x represents the user's interactive expression on the xth video type in the kth time period on the vth day; y1 represents the preset like coefficient; a v,k,x1 represents the number of likes given by the user to the xth video type in the kth time period on the vth day; y2 represents the preset comment coefficient; a v,k,x2 represents the number of comments made by the user on the xth video type in the kth time period on the vth day; y3 represents the preset forwarding coefficient; a v,k,x3 represents the number of times the user forwarded the xth video type in the kth time period on the vth day; y4 represents the preset bullet screen coefficient; a v,k,x4 represents the number of comments made by the user on the xth video type in the kth time period on the vth day; y5 represents the preset coin coefficient; a v,k,x5 It indicates the number of times the user pays coins to the xth video type in the kth time period on the vth day.
[0053] It should be noted that the coefficient of each interactive behavior comes from the statistical data of the video website. The more interactive behaviors a user has for a certain type of video in each time period of the day, the higher the user's interactive expressiveness.
[0054] In addition to the user's interactive behavior with the video, the most important factor in determining the user's preference for the video type is the time the user spends watching the video. For the film and television big data platform, the time the user spends watching the video is more important than the user's interactive behavior, because the time the user spends watching the video will be converted into the platform's daily active users, and it can also better indicate the user's preference for the video type.
[0055] According to the viewing time and total viewing time of all videos of each video type watched by users in each time period of each day, the corresponding calculation formula for the user's attention to each video type in each time period of each day is obtained as follows:
[0056]
[0057] In the formula, c v,k,x represents the user's attention to the xth video type in the kth time period on the vth day; Iv,k,x represents the number of videos of the xth video type watched by the user in the kth time period on the vth day; m v,k,x,i M represents the viewing time of the i-th video of the x-th video type in the k-th time period of the v-th day; v,k,x,i It indicates the total time that the user watches the i-th video of the x-th video type in the k-th time period on the v-th day.
[0058] It should be noted that It represents the ratio of the viewing time of the i-th video of the x-th video type to the total viewing time of the user in the k-th time period of the v-th day. The larger the ratio, the more attention the user pays to the video. It represents the ratio of the viewing time of all videos of the xth video type watched by the user in the kth time period of the vth day to the total viewing time. The larger the ratio is, the higher the user's attention to the xth video type in the kth time period of the vth day.
[0059] The product of the user's interactive expression and attention to each video type in each time period of the day is recorded as the user's liking for each video type in each time period of the day.
[0060] Step S003: According to the user's preference for each video type in each time period of all days and the time the user watches videos of all video types in each time period of every day, obtain the user's correction degree for each video type in each time period; according to the user's correction degree for each video type in each time period and the user's preference for each video type in each time period of all days, obtain the user's final preference for each video type in each time period; according to the user's final preference for each video type in each time period, obtain the user's favorite video type in each time period.
[0061] When obtaining the user's preference for each type of video, the type of videos a user likes is different in different time periods. For example, the user likes to browse news videos in the morning and food videos at noon. In the evening, when the user has more free time, he may prefer to browse complete animation videos, etc. Therefore, the user may like the same type of video to different degrees in different time periods. Therefore, it is necessary to make a more detailed correction to the user's preference for each type of video.
[0062] Within the preset total number of days, with the time period as the horizontal axis and the user's preference for the x-th video type in each time period as the vertical axis, a curve graph corresponding to the x-th video type is constructed;
[0063] According to the user's preference for each video type in each time period of all days, the time the user watches videos of all video types in each time period of each day, and the corresponding curve graph of each video type, the corresponding calculation formula for the user's correction degree for each video type in each time period is obtained as follows:
[0064]
[0065] In the formula, e k,x represents the degree of correction of the xth video type by the user in the kth time period; X represents the total number of preset days; d v,k,x N represents the user's preference for the xth video type in the kth time period on the vth day; v,k represents the sum of the time that the user watches all videos of all video types in the kth time period on the vth day; n v,k,x represents the sum of the time that the user watches all videos of the xth video type in the kth time period on the vth day; k v,x It represents the slope of the data point corresponding to the user in the kth time period on the vth day on the curve graph corresponding to the xth video type; sigmoid() is a linear normalization function that normalizes the data value to the [0,1] interval.
[0066] The average of the user's liking for each video type in each time period of all days is recorded as the user's liking for each video type in each time period;
[0067] According to the correction degree and liking degree of each video type by the user in each time period, the corresponding calculation formula for obtaining the final liking degree of each video type by the user in each time period is as follows:
[0068]
[0069] In the formula, r k,x represents the user's final preference for the xth video type in the kth time period; e k,x represents the degree of correction of the xth video type by the user in the kth time period; d k,x represents the user's preference for the xth video type in the kth time period; O represents the number of time periods in a day; d o,x Indicates the user's preference for the xth video type in the oth time period.
[0070] It should be noted that represents the average value of the user's preference for the xth video type in all time periods except the kth time period; The larger the value, the higher the user's final preference for the xth video type in the kth time period.
[0071] The video type corresponding to the maximum value of the user's final preference for each video type in each time period is recorded as the user's favorite video type in each time period; thereby, accurate video type push is performed to the user in different time periods.
[0072] So far, the present invention is completed.
[0073] In summary, in the embodiment of the present invention, the total duration of each video, the viewing duration of each video of each video type watched by the user in each time period of each day, and the user's interactive data for each video type are obtained, and the user's liking degree for each video type in each time period of each day is calculated. Combined with the time the user watches videos of all video types in each time period of each day, the user's correction degree for each video type in each time period is obtained, and then the user's final liking degree for each video type in each time period is obtained, and finally the user's favorite video type in each time period is obtained. The present invention analyzes the user's preferences for historically watched videos, customizes a personalized algorithm to recommend video types for the user, and makes more accurate video recommendations.
[0074] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for intelligent analysis and processing of data on a film and television big data platform. It is characterized in that The method comprises the following steps: Get the total duration of each video within the preset total number of days; divide each day into several time periods; get the viewing time of each video of each video type watched by the user in each time period of each day, the number of all videos of each video type watched by the user in each time period of each day, and the number of likes, comments, reposts, bullet comments, and coins for each video type by the user in each time period of each day; According to the number of likes, comments, reposts, bullet comments, and coins for each video type in each time period of the day, the user's interactive expression for each video type in each time period of the day is obtained; according to the user's interactive expression for each video type in each time period of the day, the viewing time and total viewing time of all videos of each video type, the user's liking degree for each video type in each time period of the day is obtained; Based on the user's preference for each video type in each time period of all days and the time the user spends watching videos of all video types in each time period of each day, the user's correction degree for each video type in each time period is obtained; based on the user's correction degree for each video type in each time period and the user's preference for each video type in each time period of all days, the user's final preference for each video type in each time period is obtained; based on the user's final preference for each video type in each time period, the user's favorite video type in each time period is obtained.
2. According to the method for intelligent analysis and processing of film and television big data platform data as described in claim 1, It is characterized in that According to the number of likes, comments, reposts, bullet screens, and coins for each video type by the user in each time period of each day, the specific calculation formula for the user's interactive expressiveness for each video type in each time period of each day is obtained as follows: b v,k,x =and 1 to v,k,x1 +y 2 to v,k,x2 +y 3 to v,k,x3 +y 4 to v,k,x4 +y 5 to v,k,x5 Where b v,k,x represents the user's interactive performance on the xth video type in the kth time period on the vth day; y 1 Indicates the preset like coefficient; a v,k,x1 represents the number of likes given by the user to the xth video type in the kth time period on the vth day; y 2 Indicates the preset comment coefficient; a v,k,x2 represents the number of comments made by the user on the xth video type in the kth time period on the vth day; y 3 Indicates the preset forwarding coefficient; a v,k,x3 represents the number of times the user forwarded the xth video type in the kth time period on the vth day; y 4 Indicates the preset bullet screen coefficient; a v,k,x4 represents the number of comments made by the user on the xth video type in the kth time period on the vth day; y 5 Indicates the preset coin coefficient; a v,l,x5 It indicates the number of times the user pays coins to the xth video type in the kth time period on the vth day.
3. According to the method for intelligent analysis and processing of film and television big data platform data as described in claim 1, It is characterized in that The method of obtaining the user's preference for each video type in each time period of the day according to the user's interactive expression for each video type in each time period of the day, the viewing time and the total viewing time of all videos of each video type, includes the following specific steps: According to the viewing time and total viewing time of all videos of each video type watched by users in each time period of each day, the user's attention to each video type in each time period of each day is obtained; According to the user's interactive expression and attention to each video type in each time period of the day, the user's preference degree for each video type in each time period of the day is obtained.
4. According to claim 3, a method for intelligent analysis and processing of film and television big data platform data, It is characterized in that The specific calculation formula for obtaining the user's attention to each video type in each time period of the day according to the viewing time and total viewing time of all videos of each video type watched by the user in each time period of the day is: In the formula, c v,k,x represents the user's attention to the xth video type in the kth time period on the vth day; I v,k,x represents the number of videos of the xth video type watched by the user in the kth time period on the vth day; m v,k,x,i M represents the viewing time of the i-th video of the x-th video type in the k-th time period of the v-th day; v,x,x,i It indicates the total duration of watching the i-th video of the x-th video type by the user in the k-th time period on the v-th day.
5. According to the method for intelligent analysis and processing of film and television big data platform data as described in claim 3, It is characterized in that The method of obtaining the user's preference for each video type in each time period of the day according to the user's interactive expression and attention to each video type in each time period of the day includes the following specific steps: The product of the user's interactive expression and attention to each video type in each time period of the day is recorded as the user's liking for each video type in each time period of the day.
6. According to claim 1, a method for intelligent analysis and processing of film and television big data platform data, It is characterized in that The method of obtaining the user's correction degree for each video type in each time period according to the user's preference degree for each video type in each time period of all days and the time the user watches videos of all video types in each time period of each day includes the following specific steps: Within the preset total number of days, with the time period as the horizontal axis and the user's preference for the x-th video type in each time period as the vertical axis, a curve graph corresponding to the x-th video type is constructed; According to the user's preference for each video type in each time period of all days, the time the user watches videos of all video types in each time period of each day, and the curve graph corresponding to each video type, the user's correction degree for each video type in each time period is obtained.
7. According to claim 6, a method for intelligent analysis and processing of film and television big data platform data, It is characterized in that According to the user's preference for each video type in each time period of all days, the time the user watches videos of all video types in each time period of each day, and the curve graph corresponding to each video type, the specific calculation formula for obtaining the user's correction degree for each video type in each time period is: In the formula, e k,x represents the degree of correction of the xth video type by the user in the kth time period; X represents the total number of preset days; d v,k,x N represents the user's preference for the xth video type in the kth time period on the vth day; v,k nv represents the sum of the time that the user watches all videos of all video types in the kth time period on the vth day; ,k,x represents the sum of the time that the user watches all videos of the xth video type in the kth time period on the vth day; k v,x It represents the slope of the data point corresponding to the user in the kth time period on the vth day on the curve graph corresponding to the xth video type; sigmoid() is a linear normalization function.
8. According to the method for intelligent analysis and processing of film and television big data platform data as described in claim 1, It is characterized in that The method of obtaining the final liking degree of the user for each video type in each time period according to the correction degree of the user for each video type in each time period and the liking degree of the user for each video type in each time period of all days includes the following specific steps: The average of the user's liking for each video type in each time period of all days is recorded as the user's liking for each video type in each time period; According to the correction degree and the liking degree of the user for each video type in each time period, the final liking degree of the user for each video type in each time period is obtained.
9. According to claim 8, a method for intelligent analysis and processing of film and television big data platform data, It is characterized in that The specific calculation formula for obtaining the final degree of preference of the user for each video type in each time period according to the correction degree and preference degree of the user for each video type in each time period is: In the formula, r k,x represents the user's final preference for the xth video type in the kth time period; e k,x represents the degree of correction of the xth video type by the user in the kth time period; d k,x represents the user's preference for the xth video type in the kth time period; O represents the number of time periods in a day; d o,x Indicates the user's preference for the xth video type in the oth time period.
10. According to claim 1, a method for intelligent analysis and processing of film and television big data platform data, It is characterized in that The method of obtaining the user's favorite video type in each time period according to the user's final preference degree for each video type in each time period includes the following specific steps: The video type corresponding to the maximum value of the user's final liking degree for each video type in each time period is recorded as the user's favorite video type in each time period.
Citation Information
Cited By
Family time-sharing portrait personality recommendation method and device based on real-time detection, and terminal
CN119996774A
