阅读说明：本技术 一种智能投放视频播放方法及系统 (Intelligent release video playing method and system ) 是由 梁堃 邝峰然 于 2021-09-03 设计创作，主要内容包括：本发明公开了一种智能投放视频播放方法及系统,具体包括以下步骤：步骤一、用户匹配设定；步骤二、视频初步投放；步骤三、观看记录；步骤四、完善定位投放；步骤五、投放循环完善,本发明涉及视频投放技术领域。该智能投放视频播放方法及系统,通过构建出一套完整的视频智能投放系统,对用户的行为进行分析,规划出适用于该用户的视频投放方式,提高视频投放的精准度,其中根据对用户在工作作息和生活作息时间上观看视频数据的统计,确定用户的喜好视频类型,对视频的长短进行统计整理,保证视频的长短可以有效贴合用户的生活,从而降低观看视频对用户产生的影响,保证用户的活跃度的同时,避免对用户的正常生活产生干扰。(The invention discloses an intelligent video playing method and system, which specifically comprise the following steps: step one, user matching and setting; step two, video initial release; step three, watching records; step four, perfecting positioning and putting; and fifthly, releasing circulation is complete, and the invention relates to the technical field of video releasing. According to the intelligent video playing method and system, a set of complete video intelligent delivery system is constructed, behaviors of a user are analyzed, a video delivery mode suitable for the user is planned, the video delivery accuracy is improved, the favorite video type of the user is determined according to statistics of video data watched by the user in work and rest time, the length of a video is counted and sorted, the life of the user can be effectively attached to the length of the video, the influence of watching the video on the user is reduced, and the interference on the normal life of the user is avoided while the activity of the user is guaranteed.)

1. An intelligent video playing method is characterized in that: the method specifically comprises the following steps:

step one, user matching and setting: collecting the favorite video types, work and rest information and life work and rest information filled by a user, setting the life work and rest as long-time idle time, pushing a long video to the user for playing at the long-time idle time, setting the work and rest as short-time idle time, pushing a short video to the user for playing at the short-time idle time, and completing preliminary matching with the user;

step two, video initial delivery: searching out videos of corresponding types according to the types of favorite videos filled by the user, releasing the videos of the types to the user, and sequentially and circularly releasing the videos according to the types of the favorite videos of the user in the releasing process, so that the user can watch the same number of different types of videos during initial releasing, and randomly and alternately releasing videos related to holidays and hot news when detecting that the current day is holidays or hot news exists;

step three, watching records: generating a user work and life time division table for receiving work and life information filled by a user, simultaneously acquiring the video type when the user browses the video, and marking time period information formed by the start time and the end time of the user browsed the video;

step four, perfecting positioning and putting: detecting the process of watching videos by a user in real time, caching the progress of the last video when the user quits or forcibly quits watching the videos, recording the caching times of the user watching the same video, counting the caching time information and the video type information, judging whether the user likes to browse the type of videos, namely the video types which are liked by the user and have more caching times, searching the same type of videos with corresponding lengths to be released in the time period after determining the playing length of the type of videos in the time period, screening effective browsing information according to the video playing rate after receiving the user browsing type video information collected in the step three, counting the times of the same type of videos which are effectively browsed, and generating releasing proportions of different types of videos in a mode of making ratios according to the times of the effective type of videos, releasing the video according to the proportion;

step five, the feeding circulation is complete: and in the fourth step, after video playing is set for one week, the video is transmitted to the third step as a video watching reference, and the operation of the third step and the fourth step is repeated to release the video.

2. The utility model provides an intelligence is put in video playback system which characterized in that: the video delivery system comprises an intelligent delivery system (1) and a feature video repository (2), wherein the feature video repository (2) is used for storing video resources in a classified manner according to video types, the intelligent delivery system (1) specifically comprises a user definition unit (3), a video delivery unit (4), a watching recording unit (5) and a favorite positioning perfecting unit (6), the user definition unit (3) is in butt joint with the video delivery unit (4) and is used for counting user data and delivering videos according to user data characteristics, the video delivery unit (4) is in butt joint with the watching recording unit (5) and is used for counting video data watched by a user, the watching recording unit (5) is in butt joint with the favorite positioning perfecting unit (6), and an output end of the favorite positioning perfecting unit (6) is connected with an input end of the video delivery unit (4), for further refining the type of video delivered to the user based on statistics of the video data viewed by the user.

3. The system of claim 2, wherein: the user definition unit (3) specifically comprises a user filling module (7), a long-time browsing module (8) and a free browsing module (9), wherein the user filling module (7) is used for a user to fill favorite video types, work and rest information and life and rest information;

the long-time browsing module (8) sets the daily work and rest as a long-time idle time and is used for pushing a long video to a user to play at the long-time idle time;

and the idle browsing module (9) sets the work and rest as a short-time idle time and is used for pushing a short video to a user to play at the short-time idle time.

4. The system of claim 2, wherein: the video launching unit (4) specifically comprises a mark retrieval module (10), a time length equal-proportion adjustment retrieval module (11), a hotspot random push module (12) and a feature proportion launching module (13), wherein the mark retrieval module (10) is used for retrieving related types of videos from a feature video repository (2) according to the types of favorite videos of a user;

the time length equal proportion adjustment retrieval module (11) is used for retrieving videos of the same type with corresponding lengths from the feature video storage library (2) for playing in the time period after determining the playing length of the feature videos in the corresponding time period;

the hot spot random pushing module (12) is used for detecting whether the current day is on a holiday and whether hot news exists, and randomly and alternately launching videos related to the holiday or the hot news when the current day on the holiday and the hot news are detected;

the characteristic proportion releasing module (13) is used for sequentially and circularly releasing different types according to the types of the favorite videos of the user, detecting the favorite degrees of the different types of videos of the user, receiving releasing proportion information of the different types of videos sent by the favorite positioning perfecting unit (6), and controlling the proportion when the different types of videos are released to the user.

5. The system of claim 2, wherein: the viewing recording unit (5) specifically comprises a time division module (14), a viewing characteristic marking module (15) and a time period marking module (16), wherein the time division module (14) is used for receiving the information filled in the user definition unit (3) and generating a user work and life time division table for carrying out system statistics;

the watching characteristic marking module (15) is used for collecting the video type when a user browses the video;

the time period marking module (16) is used for marking time period information formed by the starting time and the ending time of the video browsed by the user.

6. The system of claim 2, wherein: the preference positioning and perfecting unit (6) specifically comprises a forced exit progress caching module (17), a same video caching frequency recording module (18), a video preference judging and integrating counting module (19) and a video characteristic analyzing and marking module (20), wherein the forced exit progress caching module (17) caches the last video progress when a user exits or is forced to exit to watch videos in the process of watching the videos;

the same video caching frequency recording module (18) is used for recording the caching frequency when a user watches the same video;

the video preference judging, integrating and counting module (19) is used for screening out effective browsing information of the video according to the playing completion rate of the video;

the video characteristic analysis marking module (20) is used for counting the occurrence times of videos of the same type after effective browsing, generating the releasing proportion of the videos of different types in a mode of comparing the occurrence times of the videos of the effective types in a time comparison mode, and sending the releasing proportion to the video releasing unit (4) for adjusting the releasing proportion of the videos of different types.

Technical Field

The invention relates to the technical field of video delivery, in particular to an intelligent video delivery playing method and system.

Background

When a user watches videos, a plurality of recommended videos are often popped up and played automatically, and certain judgment can be performed on the user preference according to the previous watching record of the user when the device is opened, or related video recommendation is performed according to the user preference around the user according to the position of the user, or after a user address list is obtained, related videos are sorted out according to the preference of people in the user address list and are released.

The existing video intelligent delivery method and system usually enlarge the delivery of the corresponding favorite video according to whether the user likes a certain video, so that the current user is not enough to comprehensively fit the life of the user, particularly, the working rest time is limited for working personnel, the video with shorter time can be selected for watching, and the video with longer watching time is favored when the user has enough time for working and rest in holidays, and the conventional video delivery mode usually directly delivers the video related to the favorite of the user regardless of the length of the video, so that the normal life of the user is influenced for the user with poorer self-control force, therefore, the method and system for delivering the intelligent video are specially provided, and the video delivery with corresponding length is carried out according to the idle time of the user after the favorite video type of the user is screened out, effectively fit the normal life of the user.

Disclosure of Invention

Technical problem to be solved

Aiming at the defects of the prior art, the invention provides an intelligent video delivery playing method and system, which solve the problem that the normal life of a user is influenced by the user with poor self-control capability because the video related to the preference of the user is often directly delivered regardless of the length of the video in a conventional video delivery mode.

(II) technical scheme

In order to achieve the purpose, the invention is realized by the following technical scheme: an intelligent video playing method specifically comprises the following steps:

step one, user matching and setting: collecting the favorite video types, work and rest information and life work and rest information filled by a user, setting the life work and rest as long-time idle time, pushing a long video to the user for playing at the long-time idle time, setting the work and rest as short-time idle time, pushing a short video to the user for playing at the short-time idle time, and completing preliminary matching with the user;

step two, video initial delivery: searching out videos of corresponding types according to the types of favorite videos filled by the user, releasing the videos of the types to the user, and sequentially and circularly releasing the videos according to the types of the favorite videos of the user in the releasing process, so that the user can watch the same number of different types of videos during initial releasing, and randomly and alternately releasing videos related to holidays and hot news when detecting that the current day is holidays or hot news exists;

step three, watching records: generating a user work and life time division table for receiving work and life information filled by a user, simultaneously acquiring the video type when the user browses the video, and marking time period information formed by the start time and the end time of the user browsed the video;

step four, perfecting positioning and putting: detecting the process of watching videos by a user in real time, caching the progress of the last video when the user quits or forcibly quits watching the videos, recording the caching times of the user watching the same video, counting the caching time information and the video type information, judging whether the user likes to browse the type of videos, namely the video types which are liked by the user and have more caching times, searching the same type of videos with corresponding lengths to be released in the time period after determining the playing length of the type of videos in the time period, screening effective browsing information according to the video playing rate after receiving the user browsing type video information collected in the step three, counting the times of the same type of videos which are effectively browsed, and generating releasing proportions of different types of videos in a mode of making ratios according to the times of the effective type of videos, releasing the video according to the proportion;

step five, the feeding circulation is complete: and in the fourth step, after video playing is set for one week, the video is transmitted to the third step as a video watching reference, and the operation of the third step and the fourth step is repeated to release the video.

By adopting the technical scheme, the favorite video types of the users are determined according to the statistics of the watching video data of the users in the working and living work and rest time, the length of the videos is counted and sorted, the length of the videos can be guaranteed to be effectively fit with the life of the users, the influence of watching the videos on the users is reduced, and the activity of the users is guaranteed.

The invention also discloses an intelligent release video playing system, which is a video release system consisting of an intelligent release system and a characteristic video storage library, the intelligent delivery system specifically comprises a user definition unit, a video delivery unit, a watching recording unit and a favorite positioning perfecting unit, wherein the user definition unit is in butt joint with the video delivery unit, used for counting user data and carrying out video delivery according to the characteristics of the user data, the video delivery unit is butted with the watching recording unit, the device is used for realizing the statistics of video data watched by a user, the watching recording unit is butted with the favorite positioning perfecting unit, the output end of the favorite positioning perfecting unit is connected with the input end of the video releasing unit, for further refining the type of video delivered to the user based on statistics of the video data viewed by the user.

By adopting the technical scheme, a set of complete video intelligent delivery system is constructed, the behaviors of the user are analyzed, a video delivery mode suitable for the user is planned, and the video delivery accuracy is improved.

The invention is further configured to: the user definition unit specifically comprises a user filling module, a long-time browsing module and an idle browsing module, wherein the user filling module is used for filling favorite video types, work and rest information and life and rest information for a user;

the long-time browsing module sets the daily work and rest as a long-time idle time and is used for pushing a long video to a user to play at the long-time idle time;

and the idle browsing module sets the work and rest as a short-time idle time and is used for pushing the short video to the user to play at the short-time idle time.

By adopting the technical scheme, the initial form of the user behavior is defined, and the release reference is provided for the initial video release.

The invention is further configured to: the video releasing unit specifically comprises a mark retrieving module, a time length equal proportion adjusting retrieving module, a hotspot random pushing module and a feature proportion releasing module, wherein the mark retrieving module is used for retrieving related types of videos from a feature video repository according to the types of the favorite videos of a user;

the time length equal proportion adjustment retrieval module is used for retrieving videos of the same type with corresponding lengths from the feature video storage library for playing in the time period after determining the playing length of the feature videos in the corresponding time period;

the hot spot random pushing module is used for detecting whether the current day is on a holiday and whether hot news exists, and randomly and alternately launching videos related to the holiday or the hot news when the current day is detected to be on the holiday and the hot news exists;

the characteristic proportion releasing module is used for sequentially and circularly releasing different types according to the types of the favorite videos of the user, detecting the favorite degrees of the user on the different types of videos, receiving releasing proportion information of the different types of videos sent by the favorite positioning perfecting unit, and controlling the proportion when the different types of videos are released to the user.

By adopting the technical scheme, the videos with different lengths can be retrieved under the limitation of the time period, the retrieval of the videos with different types is realized, and the retrieved videos are released.

The invention is further configured to: the viewing recording unit specifically comprises a time division module, a viewing characteristic marking module and a time period marking module, wherein the time division module is used for receiving the information filled in the user definition unit, generating a user work and life time division table and carrying out system statistics;

the watching characteristic marking module is used for collecting the video type when a user browses the video;

the time period marking module is used for marking time period information formed by the starting time and the ending time of the video browsed by the user.

By adopting the technical scheme, the statistics of the video type information and the time information when the user browses the video is realized.

The invention is further configured to: the preference positioning and perfecting unit specifically comprises a forced exit progress caching module, a same video caching frequency recording module, a video preference judging and integrating counting module and a video characteristic analyzing and marking module, wherein the forced exit progress caching module caches the last video progress when a user exits or forcibly exits to watch a video in the process of watching the video;

the same video caching frequency recording module is used for recording the caching frequency when a user watches the same video;

the video preference judging, integrating and counting module is used for screening out effective browsing information of the video according to the playing completion rate of the video;

the video characteristic analysis marking module is used for counting the occurrence times of the same type of videos after effective browsing, generating the releasing proportion of different types of videos in a mode of comparing the occurrence times of the effective types of videos in a mode of making a ratio of the occurrence times, and sending the releasing proportion to the video releasing unit for adjusting the releasing proportion of the different types of video screens.

By adopting the technical scheme, whether the user likes the video of the type is judged according to the caching times, namely the caching times are more, the user likes more, and the releasing proportion of different types of videos in the favorite video is determined according to the watching amount of the favorite video of the user, so that the video is enabled to be better attached to the aesthetic sense of the user, and the effect of reducing the aesthetic fatigue is achieved.

(III) advantageous effects

The invention provides an intelligent video delivery playing method and system. The method has the following beneficial effects:

(1) according to the intelligent video delivery playing method and system, a set of complete video intelligent delivery system is constructed, behaviors of a user are analyzed, a video delivery mode suitable for the user is planned, the video delivery accuracy is improved, favorite video features of the user are determined according to statistics of video data watched by the user in work and rest time, the length of a video is counted and sorted, the life of the user can be effectively fitted with the length of the video, the influence of watching the video on the user is reduced, and the activity of the user is guaranteed.

(2) According to the method and the system for intelligently releasing the video, the caching times of the user watching the same video are counted through the same video caching time recording module, the user can be expressed about the favorite of the user on the type of video, namely the caching times are more, the user likes more, the time use of the user can be judged, namely the caching time is more, the user time is about scattered, the short-duration videos of the same type can be released, the user involvement on the same video is reduced as much as possible, and the interference on the normal life of the user is avoided.

(3) According to the intelligent video playing method and system, the video playing proportion of different types of favorite videos is determined according to the watching amount of the favorite videos of the user, so that the videos are enabled to better fit the aesthetic feeling of the user, and the effect of reducing the aesthetic fatigue is achieved.

Drawings

FIG. 1 is a schematic block diagram of the system of the present invention;

FIG. 2 is a system schematic block diagram of a user-defined element of the present invention;

FIG. 3 is a schematic block diagram of a system of a video delivery unit of the present invention;

FIG. 4 is a schematic block diagram of a system for viewing a recording unit in accordance with the present invention;

fig. 5 is a schematic block diagram of a system for a preference location refinement unit according to the present invention.

In the figure, 1, an intelligent delivery system; 2. a feature video repository; 3. a user definition unit; 4. a video delivery unit; 5. a viewing recording unit; 6. a preference positioning perfecting unit; 7. a user population module; 8. a long-time browsing module; 9. an idle browsing module; 10. a tag retrieval module; 11. a retrieval module for adjusting the time length in equal proportion; 12. a hotspot random push module; 13. a characteristic proportion releasing module; 14. a time division module; 15. a viewing feature labeling module; 16. a time period marking module; 17. a forced exit progress cache module; 18. the same video caching frequency recording module; 19. a video preference determination integration statistic module; 20. and a video feature analysis marking module.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Referring to fig. 1-5, an embodiment of the present invention provides a technical solution: an intelligent video playing method specifically comprises the following steps:

step one, user matching and setting: the method comprises the steps of collecting favorite video types, work and rest information and life and rest information filled by a user, further explaining, wherein the video types include but are not limited to comedies, action pieces and health-care pieces, setting the life and rest as long-time idle time, pushing long videos to the user to play at the long-time idle time, setting the work and rest as short-time idle time, pushing short videos to the user to play at the short-time idle time, and completing preliminary matching with the user;

step two, video initial delivery: according to the favorite video types filled by the user, retrieving videos of corresponding types, and releasing the videos of the types to the user, wherein in the releasing process, the videos are released circularly according to the types of the favorite videos of the user in sequence, and further, the specific mode of circular releasing comprises the following steps: after a user selects comedy, action and service slice, firstly putting in a comedy, then putting in an action, putting in a service slice, and then putting in a comedy, wherein the cycle is adopted, so that when the initial putting is ensured, the user can watch different types of videos with the same quantity, and when the video is put in a festival or has hot news, the videos related to the festival and the hot news are randomly and alternately put in during the video putting, further, after the festival is detected, a video related to the festival is randomly put in and inserted between the put comedy and the action, and similarly, the hot news related video is also put in the mode and is used as a plug-in video to be put in under the condition that the integral putting of the video is not influenced;

step three, watching records: generating a user work and life time division table for receiving work and life information filled by a user, simultaneously acquiring the video type when the user browses the video, and marking time period information formed by the start time and the end time of the user browsed the video;

step four, perfecting positioning and putting: detecting the process of watching videos by a user in real time, caching the progress of the last video when the user quits or forcibly quits watching the videos, recording the caching times of the user watching the same video, counting the caching time information and the video type information, judging whether the user likes to browse the type of videos, namely the video types which are liked by the user and have more caching times, searching the same type of videos with corresponding lengths to be released in the time period after determining the playing length of the type of videos in the time period, screening effective browsing information according to the video playing rate after receiving the user browsing type video information collected in the step three, counting the times of the same type of videos which are effectively browsed, and generating releasing proportions of different types of videos in a mode of making ratios according to the times of the effective type of videos, releasing the video according to the proportion;

step five, the feeding circulation is complete: and in the fourth step, after video playing is set for one week, the video is transmitted to the third step as a video watching reference, and the operation of the third step and the fourth step is repeated to release the video.

The above-mentioned intelligent delivery video playing system, as shown in fig. 1, is a video delivery system composed of an intelligent delivery system 1 and a feature video repository 2, where the feature video repository 2 is used to store video resources in a classified manner according to video types, the intelligent delivery system 1 specifically includes a user definition unit 3, a video delivery unit 4, a watching recording unit 5, and a favorite positioning completion unit 6, specifically, as shown in fig. 2, the user definition unit 3 specifically includes a user filling module 7, a long-time browsing module 8, and a casual browsing module 9, and the user filling module 7 is used for a user to fill in favorite video types, work and rest information;

the long-time browsing module 8 sets the daily work and rest as a long-time idle time, and is used for pushing a long video to a user to play at the long-time idle time;

the idle browsing module 9 sets the work and rest as a short-time idle time, and is used for pushing the short video to the user for playing at the short-time idle time.

The user defining unit 3 is butted with the video releasing unit 4, and is configured to count user data, release videos according to user data characteristics, and implement user matching setting, specifically, as shown in fig. 3, the video releasing unit 4 specifically includes a mark retrieving module 10, a duration equal-proportion adjustment retrieving module 11, a hotspot random pushing module 12, and a characteristic proportion releasing module 13, where the mark retrieving module 10 is configured to retrieve videos of related types from the characteristic video repository 2 according to the types of favorite videos of the user;

the time length equal proportion adjustment retrieval module 11 is used for retrieving videos of the same type with corresponding lengths from the feature video storage 2 for playing in the time period after determining the playing length of the feature video in the corresponding time period;

the hot spot random pushing module 12 is configured to detect whether the current day is a holiday and whether hot news exists, and randomly insert and release videos related to the holiday or the hot news when the current day is detected to be the holiday and the hot news exists;

the characteristic proportion releasing module 13 is used for sequentially and circularly releasing different types according to the types of the favorite videos of the user, detecting the favorite degrees of the different types of videos of the user, receiving releasing proportion information of the different types of videos sent by the favorite positioning perfecting unit 6, and controlling the proportion when the different types of videos are released to the user.

The video delivery unit 4 is butted with the watching recording unit 5, and is used for counting video data watched by a user and delivering the video data after initial video delivery and perfect positioning, specifically, as shown in fig. 4, the watching recording unit 5 specifically includes a time division module 14, a watching feature marking module 15 and a time period marking module 16, the time division module 14 is used for receiving information filled in the user definition unit 3, and generating a user work and life time division table for performing system statistics;

the viewing characteristic marking module 15 is used for collecting the video type when the user browses the video;

the time period marking module 16 is used for marking the time period information formed by the start time and the end time of the video browsed by the user.

The watching recording unit 5 is connected with the fond positioning perfecting unit 6 in a butt joint mode, the output end of the fond positioning perfecting unit 6 is connected with the input end of the video releasing unit 4 and used for further perfecting the type of videos released by the user according to statistics of video data watched by the user, specifically, as shown in the attached drawing 5, the fond positioning perfecting unit 6 specifically comprises a forced quitting progress caching module 17, a same video caching frequency recording module 18, a video fond judging and integrating statistics module 19 and a video characteristic analysis marking module 20, and the forced quitting progress caching module 17 caches the last occurring video progress when quitting or forcibly quitting watching videos occurs in the process of watching videos by the user;

the same video caching frequency recording module 18 is used for recording the caching frequency when the user watches the same video;

the video preference judging, integrating and counting module 19 is used for screening out effective browsing information of the video according to the playing completion rate of the video;

the video feature analysis marking module 20 is configured to count the occurrence times of the same type of video after effective browsing, generate the release ratios of different types of videos in a manner of comparing the occurrence times of the effective type of video with the occurrence times, and send the release ratios to the video release unit 4 for adjusting the release ratios of the different types of video.

Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.