CN104918118A - Video recommendation method and device based on historic information - Google Patents
Video recommendation method and device based on historic information Download PDFInfo
- Publication number
- CN104918118A CN104918118A CN201510341831.XA CN201510341831A CN104918118A CN 104918118 A CN104918118 A CN 104918118A CN 201510341831 A CN201510341831 A CN 201510341831A CN 104918118 A CN104918118 A CN 104918118A
- Authority
- CN
- China
- Prior art keywords
- video
- feature
- type
- videos
- demand
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4662—Learning process for intelligent management, e.g. learning user preferences for recommending movies characterized by learning algorithms
- H04N21/4665—Learning process for intelligent management, e.g. learning user preferences for recommending movies characterized by learning algorithms involving classification methods, e.g. Decision trees
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/462—Content or additional data management e.g. creating a master electronic programme guide from data received from the Internet and a Head-end or controlling the complexity of a video stream by scaling the resolution or bit-rate based on the client capabilities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4667—Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
本发明公开了一种基于历史信息的视频推荐方法和装置，其中所述装置包括信息获取模块；特征需求强度获取模块；以及视频推荐模块；其中，所述特征需求强度获取模块包括特征组确定子模块，用于根据该类型视频中各视频的特征，确定该类型视频中的各特征组；以及特征分组子模块，用于将该类型视频中的各视频根据其特征分入各特征组中。根据本发明的实施例，可以简单有效地提供多类型和多特征的视频推荐结果，满足了用户复杂的视频推荐需求。
The invention discloses a video recommendation method and device based on historical information, wherein the device includes an information acquisition module; a feature demand strength acquisition module; and a video recommendation module; wherein the feature demand strength acquisition module includes a feature group determiner The module is used to determine each feature group in this type of video according to the features of each video in this type of video; and the feature grouping sub-module is used to classify each video in this type of video into each feature group according to its features. According to the embodiments of the present invention, multi-type and multi-feature video recommendation results can be provided simply and effectively, satisfying complex video recommendation requirements of users.
Description
本发明专利申请是申请日为2012年10月24日、申请号为201210408686.9、名称为“基于历史信息的视频推荐方法和装置”的中国发明专利申请的分案申请。The patent application of the present invention is a divisional application of the Chinese invention patent application with the filing date of October 24, 2012, the application number of 201210408686.9, and the title of "Video Recommendation Method and Device Based on Historical Information".
技术领域technical field
本发明涉及在线视频技术，尤其涉及一种基于历史信息的视频推荐方法和装置。The present invention relates to online video technology, in particular to a video recommendation method and device based on historical information.
背景技术Background technique
在线视频推荐是视频网站帮助用户查找并观看某个特定领域视频的方法和工具。相对于传统的视频目录浏览方式或者视频搜索方式，视频推荐能够在用户不确定合适的搜索词的情况下，通过分析用户历史行为，发现用户需求的特定领域，在该领域内进行推荐，避免了搜索词的输入和层次目录的多次点击过程，使得查找并观看某个特定类型的视频更加简单容易。Online video recommendation is a method and tool for video websites to help users find and watch videos in a specific field. Compared with the traditional video catalog browsing method or video search method, video recommendation can find the specific field of user needs by analyzing the user's historical behavior when the user is not sure about the appropriate search term, and make recommendations in this field, avoiding the The multi-click process of entering search terms and hierarchical directories makes finding and watching a particular type of video simple and easy.
现有的视频推荐技术，主要包括两种方法—基于视频协同过滤推荐和基于用户协同过滤推荐。前者通过计算视频和视频的相似度，将与观影记录视频最相似的视频推荐给用户。而后者则是基于观影记录，计算用户相似度，将相似的用户最近看过的视频推荐给用户。这两种方式默认都是基于用户的全部观影记录进行分析，返回的结果是与所有历史视频均相似的视频，对于喜好比较单一的用户，推荐结果较好。例如用户看了一部或多部动作片，推出最近最热的动作片，用户感受会比较好。Existing video recommendation technologies mainly include two methods—video-based collaborative filtering recommendation and user-based collaborative filtering recommendation. The former recommends the most similar video to the user by calculating the similarity between the video and the video. The latter is based on viewing records, calculates user similarity, and recommends videos that similar users have recently watched to users. By default, these two methods are based on the analysis of all the user's viewing records, and the returned results are videos that are similar to all historical videos. For users with a single preference, the recommendation results are better. For example, if a user has watched one or more action movies and releases the hottest action movie recently, the user experience will be better.
图1示出了现有技术(CN102306178A，“视频推荐方法及装置”)的视频推荐方法的流程图。如图1所示，在现有技术中，(1)从用户日志数据库提取每一个COOKIE观看的VIDEO(视频)作为训练样本。(2)计算所述训练样本中所有COOKIE与VIDEO之间的转移概率对，得到COOKIE到VIDEO的转移概率矩阵和VIDEO到COOKIE的转移概率矩阵。(3)根据所述COOKIE到VIDEO的转移概率矩阵和VIDEO到COOKIE矩阵，得到VIDEO之间的转移概率矩阵。(4)根据VIDEO之间的转移概率矩阵得到推荐模型，并嵌入所述用户视频搜索系统以向用户返回推荐结果。Fig. 1 shows a flowchart of a video recommendation method in the prior art (CN102306178A, "Video Recommendation Method and Device"). As shown in Figure 1, in the prior art, (1) extract the VIDEO (video) watched by each COOKIE as a training sample from the user log database. (2) Calculate transition probability pairs between all COOKIEs and VIDEOs in the training samples, and obtain a transition probability matrix from COOKIE to VIDEO and a transition probability matrix from VIDEO to COOKIE. (3) Obtain a transition probability matrix between VIDEOs according to the transition probability matrix from COOKIE to VIDEO and the matrix from VIDEO to COOKIE. (4) Obtain a recommendation model according to the transition probability matrix between VIDEOs, and embed it into the user video search system to return recommendation results to the user.
现有技术方案可以满足视频类型和特征有单一喜好的用户需求。但是随着互联网视频网站的发展和用户上网观看视频的行为增多，用户对观看视频类型和特征的需求更为多样，满足全部类型和特征的视频将不存在或者质量较差，很可能是包含较多特征但是没有一个优秀特征的视频。The prior art solution can meet the needs of users who have a single preference for video types and features. However, with the development of Internet video sites and the increase in the behavior of users watching videos on the Internet, users have more diverse needs for viewing video types and characteristics. Videos that meet all types and characteristics will not exist or the quality will be poor. A video with many features but not one outstanding feature.
在现有技术方案中，多类型的视频推荐无法得到满足：视频类型是视频资源的一个强特征，不同类型的视频推荐用户感受往往比较差。对一个准备周末花好几个小时来看爱情韩剧的用户，推荐一个只有1.5小时的爱情电影，用户感受不太好，同样给喜欢看体育短视频的用户推荐1.5小时以上的体育电影显然也不满足用户需求。另外，多特征的视频推荐也无法得到满足：相似或相同特征的视频是适合联合推荐的，不同特征的视频则不适合一起推荐。如“无间道1”、“无间道2”适合一起推荐，推荐“无间道3”、“窃听风云”比较好；而“笔仙”、“桃姐”则不合适一起推荐。In existing technical solutions, multi-type video recommendation cannot be satisfied: video type is a strong feature of video resources, and different types of video recommendation users often experience poor experience. For a user who is going to spend several hours watching romantic Korean dramas on weekends, recommending a 1.5-hour love movie is not very good for the user. Similarly, recommending sports movies of more than 1.5 hours to users who like to watch short sports videos is obviously not satisfactory. User needs. In addition, multi-feature video recommendation cannot be satisfied: videos with similar or same features are suitable for joint recommendation, while videos with different features are not suitable for joint recommendation. For example, "Infernal Affairs 1" and "Infernal Affairs 2" are suitable for recommendation together, and "Infernal Affairs 3" and "Eavesdropping Fengyun" are better recommended; while "Pen Xian" and "Sister Tao" are not suitable for recommendation together.
发明内容Contents of the invention
鉴于上述问题，提出了本发明，以便提供一种克服上述问题或者至少部分地解决上述问题的基于历史信息的视频推荐方法和装置。In view of the above problems, the present invention is proposed to provide a video recommendation method and device based on historical information that overcome the above problems or at least partially solve the above problems.
依据本发明的一个方面，提供了一种基于历史信息的视频推荐方法，包括以下步骤：According to one aspect of the present invention, a video recommendation method based on historical information is provided, comprising the following steps:
获取用户的视频观看记录信息；Obtain the user's video viewing record information;
根据所述视频观看记录信息，计算用户观看过的各类型视频的类型需求强度；According to the video viewing record information, calculate the type demand intensity of each type of video that the user has watched;
对于每一类型视频，根据视频特征进行分组，并且获取各特征组的特征需求强度；以及For each type of video, group according to video features, and obtain the feature demand strength of each feature group; and
基于所述类型需求强度和/或所述特征需求强度，向用户推荐视频。Based on the genre demand strength and/or the feature demand strength, videos are recommended to the user.
根据本发明的实施例，所述根据所述视频观看记录信息计算用户观看过的各类型视频的类型需求强度的步骤包括：According to an embodiment of the present invention, the step of calculating the type demand intensity of each type of video watched by the user according to the video viewing record information includes:
统计用户观看过的视频的类型；Statistics on the types of videos watched by users;
对于每一类型视频，根据该类型视频数量和所有视频总数量，计算该类型视频的内容需求强度；For each type of video, calculate the content demand intensity of this type of video based on the number of videos of this type and the total number of all videos;
根据该类型视频的观看时间在所有视频的观看时间中所处的时间先后位置，计算该类型视频的时间需求强度；以及Calculate the time demand intensity of this type of video according to the chronological position of the viewing time of this type of video among the viewing times of all videos; and
基于所述内容需求强度和所述时间需求强度，计算该类型视频的类型需求强度。Based on the content demand strength and the time demand strength, the genre demand strength of the type of video is calculated.
根据本发明的实施例，在所述基于所述内容需求强度和所述时间需求强度、计算该类型视频的类型需求强度的步骤中，基于以下公式计算所述类型需求强度：According to an embodiment of the present invention, in the step of calculating the genre demand strength of this type of video based on the content demand strength and the time demand strength, the type demand strength is calculated based on the following formula:
类型需求强度＝a×内容需求强度+(1-a)×时间需求强度，其中a是预先定义的常数。Type requirement strength=a×content requirement strength+(1−a)×temporal requirement strength, where a is a predefined constant.
根据本发明的实施例，所述对于每一类型视频、根据视频特征进行分组的步骤包括：According to an embodiment of the present invention, the step of grouping according to video features for each type of video includes:
根据该类型视频中各视频的特征，确定该类型视频中的各特征组；以及Determining groups of features in videos of that type based on features of videos in that type of video; and
将该类型视频中的各视频根据其特征分入各特征组中。Each video in this type of video is divided into each feature group according to its feature.
根据本发明的实施例，所述根据该类型视频中各视频的特征、确定该类型视频中的各特征组的步骤是利用Canopy聚类算法执行的，包括以下步骤：According to an embodiment of the present invention, the step of determining each feature group in this type of video according to the characteristics of each video in this type of video is performed using the Canopy clustering algorithm, including the following steps:
设置第一距离阈值和第二距离阈值，其中所述第一距离阈值小于所述第二距离阈值；setting a first distance threshold and a second distance threshold, wherein the first distance threshold is smaller than the second distance threshold;
将特征差异小于所述第一距离阈值的视频分入相同的特征组中；Classifying videos with feature differences smaller than the first distance threshold into the same feature group;
将与一特征组的特征差异小于所述第二距离阈值、但大于所述第一距离阈值的视频分入该特征组，并且另外分入单独的特征组中；以及sorting videos whose features differ from a feature group by less than said second distance threshold but greater than said first distance threshold into that feature group, and additionally into a separate feature group; and
根据特征组中的视频，计算各特征组的中心特征。Based on the videos in the feature group, calculate the central feature of each feature group.
根据本发明的实施例，所述将该类型视频中的各视频根据其特征分入各特征组中的步骤是利用K-Means聚类算法执行的，包括以下步骤：According to an embodiment of the present invention, the step of classifying each video in this type of video into each feature group according to its characteristics is performed using the K-Means clustering algorithm, including the following steps:
计算所述各视频与各特征组的中心特征的差异；Calculate the difference between each video and the central feature of each feature group;
将所述各视频分入与其差异最小的特征组中；Classify each video into the feature group with the smallest difference;
根据特征组中的视频，重新计算各特征组的中心特征；以及recalculate the central feature of each feature group based on the videos in the feature group; and
重复执行上述步骤，直到所述各特征组的中心特征与前一次计算的各特征组的中心特征之间的差异小于预先定义的阈值为止。The above steps are repeatedly executed until the difference between the center feature of each feature group and the center feature of each feature group calculated last time is smaller than a predefined threshold.
根据本发明的实施例，特征组的特征需求强度是根据该特征组中的视频的观看时间在所有特征组的视频的观看时间中所处的时间先后位置确定的。According to an embodiment of the present invention, the feature demand intensity of a feature group is determined according to the chronological positions of the viewing time of videos in the feature group among the viewing times of videos in all feature groups.
根据本发明的实施例，特征组的特征需求强度是根据该特征组中最新观看的视频的观看时间在各特征组中最新观看的视频的观看时间中所处的时间先后位置确定的。According to an embodiment of the present invention, the feature demand intensity of a feature group is determined according to the chronological position of the viewing time of the latest watched video in the feature group among the viewing times of the latest watched video in each feature group.
根据本发明的实施例，所述基于所述类型需求强度和/或所述特征需求强度、向用户推荐视频的步骤包括：According to an embodiment of the present invention, the step of recommending videos to the user based on the type demand strength and/or the feature demand strength includes:
按照类型需求强度从高到低的顺序，向用户推荐各类型的视频；以及Recommending various types of videos to the user in descending order of type demand intensity; and
对于每一类型视频，按照特征需求强度从高到低的顺序，向用户推荐各特征组的视频。For each type of video, videos of each feature group are recommended to users in order of feature demand strength from high to low.
根据本发明的实施例，所述基于所述类型需求强度和/或所述特征需求强度、向用户推荐视频的步骤还包括：According to an embodiment of the present invention, the step of recommending videos to the user based on the type demand strength and/or the feature demand strength further includes:
响应于用户更换视频类型的请求，切换向用户推荐的视频类型；和/或In response to the user's request to change the video type, switch the type of video recommended to the user; and/or
响应于用户更换视频特征组的请求，切换向用户推荐的视频特征组。In response to the user's request to change the video feature set, the video feature set recommended to the user is switched.
根据本发明的实施例，所述视频观看记录信息包含在用户的Cookie文件中。According to an embodiment of the present invention, the video watching record information is included in the user's cookie file.
依据本发明的另一方面，提供了一种基于历史信息的视频推荐装置，包括：According to another aspect of the present invention, a video recommendation device based on historical information is provided, including:
信息获取模块，用于获取用户的视频观看记录信息；An information acquisition module, configured to acquire the user's video viewing record information;
类型需求强度计算模块，用于根据所述视频观看记录信息，计算用户观看过的各类型视频的类型需求强度；Type demand strength calculation module, used to calculate the type demand strength of each type of video that the user has watched according to the video viewing record information;
特征需求强度获取模块，用于对于每一类型视频，根据视频特征进行分组，并且获取各特征组的特征需求强度；以及A feature demand strength acquisition module, configured to group each type of video according to video features, and obtain the feature demand strength of each feature group; and
视频推荐模块，用于基于所述类型需求强度和/或所述特征需求强度，向用户推荐视频。A video recommending module, configured to recommend videos to users based on the genre demand strength and/or the feature demand strength.
根据本发明的实施例，所述类型需求强度计算模块包括：According to an embodiment of the present invention, the type demand intensity calculation module includes:
类型统计子模块，用于统计用户观看过的视频的类型；The type statistics sub-module is used to count the types of videos watched by users;
内容需求强度计算子模块，用于对于每一类型视频，根据该类型视频数量和所有视频总数量，计算该类型视频的内容需求强度；The content demand intensity calculation sub-module is used for calculating the content demand intensity of this type of video according to the number of videos of this type and the total number of all videos for each type of video;
时间需求强度计算子模块，用于根据该类型视频的观看时间在所有视频的观看时间中所处的时间先后位置，计算该类型视频的时间需求强度；以及The time demand intensity calculation sub-module is used to calculate the time demand intensity of this type of video according to the time sequence position of the viewing time of this type of video in the viewing time of all videos; and
类型需求强度计算子模块，用于基于所述内容需求强度和所述时间需求强度，计算该类型视频的类型需求强度。The type demand strength calculation sub-module is used to calculate the type demand strength of this type of video based on the content demand strength and the time demand strength.
根据本发明的实施例，所述类型需求强度计算子模块(203d)基于以下公式计算所述类型需求强度：According to an embodiment of the present invention, the type demand strength calculation submodule (203d) calculates the type demand strength based on the following formula:
类型需求强度＝a×内容需求强度+(1-a)×时间需求强度，其中a是预先定义的常数。Type requirement strength=a×content requirement strength+(1−a)×temporal requirement strength, where a is a predefined constant.
根据本发明的实施例，所述特征需求强度获取模块包括：According to an embodiment of the present invention, the feature demand intensity acquisition module includes:
特征组确定子模块，用于根据该类型视频中各视频的特征，确定该类型视频中的各特征组；以及The feature group determination submodule is used to determine each feature group in this type of video according to the features of each video in this type of video; and
特征分组子模块，用于将该类型视频中的各视频根据其特征分入各特征组中。The feature grouping sub-module is used to classify each video in the type of video into each feature group according to its feature.
根据本发明的实施例，所述特征组确定子模块利用Canopy聚类算法，根据该类型视频中各视频的特征，确定该类型视频中的各特征组，其中所述特征组确定子模块：According to an embodiment of the present invention, the feature group determination submodule utilizes the Canopy clustering algorithm to determine each feature group in this type of video according to the characteristics of each video in this type of video, wherein the feature group determination submodule:
设置第一距离阈值和第二距离阈值，其中所述第一距离阈值小于所述第二距离阈值；setting a first distance threshold and a second distance threshold, wherein the first distance threshold is smaller than the second distance threshold;
将特征差异小于所述第一距离阈值的视频分入相同的特征组中；Classifying videos with feature differences smaller than the first distance threshold into the same feature group;
将与一特征组的特征差异小于所述第二距离阈值、但大于所述第一距离阈值的视频分入该特征组，并且另外分入单独的特征组中；以及sorting videos whose features differ from a feature group by less than said second distance threshold but greater than said first distance threshold into that feature group, and additionally into a separate feature group; and
根据特征组中的视频，计算各特征组的中心特征。Based on the videos in the feature group, calculate the central feature of each feature group.
根据本发明的实施例，所述特征分组子模块利用K-Means聚类算法将该类型视频中的各视频根据其特征分入各特征组中，其中所述特征分组子模块(205b)：According to an embodiment of the present invention, the feature grouping submodule utilizes the K-Means clustering algorithm to divide each video in this type of video into each feature group according to its characteristics, wherein the feature grouping submodule (205b):
计算所述各视频与各特征组的中心特征的差异；Calculate the difference between each video and the central feature of each feature group;
将所述各视频分入与其差异最小的特征组中；Classify each video into the feature group with the smallest difference;
根据特征组中的视频，重新计算各特征组的中心特征；以及recalculate the central feature of each feature group based on the videos in the feature group; and
重复执行上述步骤，直到所述各特征组的中心特征与前一次计算的各特征组的中心特征之间的差异小于预先定义的阈值为止。The above steps are repeatedly executed until the difference between the center feature of each feature group and the center feature of each feature group calculated last time is smaller than a predefined threshold.
根据本发明的实施例，所述特征需求强度获取模块根据该特征组中的视频的观看时间在所有特征组的视频的观看时间中所处的时间先后位置，确定特征组的特征需求强度。According to an embodiment of the present invention, the feature demand strength acquisition module determines the feature demand strength of the feature group according to the chronological positions of the viewing time of the videos in the feature group among the viewing times of videos of all feature groups.
根据本发明的实施例，所述特征需求强度获取模块根据该特征组中最新观看的视频的观看时间在各特征组中最新观看的视频的观看时间中所处的时间先后位置，确定特征组的特征需求强度。According to an embodiment of the present invention, the feature demand intensity acquisition module determines the feature group according to the viewing time of the latest watched video in the feature group and the time sequence position of the latest watched video in each feature group. Characteristic demand strength.
根据本发明的实施例，其中所述视频推荐模块包括：According to an embodiment of the present invention, wherein the video recommendation module includes:
类型推荐子模块，用于按照类型需求强度从高到低的顺序，向用户推荐各类型的视频；以及A type recommendation sub-module is used to recommend various types of videos to users in order of type demand intensity from high to low; and
特征组推荐子模块，用于对于每一类型视频，按照特征需求强度从高到低的顺序，向用户推荐各特征组的视频。The feature group recommendation sub-module is used for recommending videos of each feature group to the user in order of feature demand intensity from high to low for each type of video.
根据本发明的实施例，所述视频推荐模块还包括：According to an embodiment of the present invention, the video recommendation module also includes:
类型切换子模块，用于响应于用户更换视频类型的请求，切换向用户推荐的视频类型；和/或A type switching submodule, configured to switch the type of video recommended to the user in response to the user's request for changing the type of video; and/or
特征组切换子模块，用于响应于用户更换视频特征组的请求，切换向用户推荐的视频特征组。The feature group switching sub-module is used to switch the video feature group recommended to the user in response to the user's request to change the video feature group.
根据本发明的实施例，所述视频观看记录信息包含在用户的Cookie文件中。According to an embodiment of the present invention, the video watching record information is included in the user's cookie file.
本发明提供了基于历史信息的视频推荐方法和装置。根据本发明的实施例，根据用户的视频观看记录信息，计算各类型视频的类型需求强度，将每一类型视频中的视频分入各特征组，并获取各特征组的特征需求强度，最后基于类型需求强度和/或特征需求强度向用户推荐视频，可以简单有效地提供多类型和多特征的视频推荐结果，满足了用户复杂的视频推荐需求。相对于传统的视频推荐方法，本发明能够计算用户对于不同类型视频的类型需求强度，并据此以不同的优先级推荐电影、电视剧、动漫剧、综艺节目、体育节目等各种类型的视频，并且根据用户的特征需求强度，以不同的优先级推荐特定视频类型下的不同特征的视频，如电影类型中的爱情片、科幻片、战争片，电视剧类型中家庭剧、历史剧等等。另外，还可以响应于用户的请求，切换向用户推荐的视频类型和/或视频特征组。The invention provides a video recommendation method and device based on historical information. According to an embodiment of the present invention, according to the user's video viewing record information, the type demand strength of each type of video is calculated, the videos in each type of video are divided into each feature group, and the feature demand strength of each feature group is obtained, and finally based on The genre demand strength and/or feature demand strength recommend videos to users, which can simply and effectively provide multi-type and multi-feature video recommendation results, and meet the complex video recommendation needs of users. Compared with the traditional video recommendation method, the present invention can calculate the type demand intensity of users for different types of videos, and accordingly recommend various types of videos such as movies, TV dramas, animation dramas, variety shows, sports programs, etc. with different priorities, And according to the strength of the user's characteristic needs, videos with different characteristics under specific video types are recommended with different priorities, such as romance films, science fiction films, and war films in movie types, family dramas, historical dramas in TV series, and so on. In addition, the video type and/or video feature group recommended to the user may also be switched in response to the user's request.
上述说明仅是本发明技术方案的概述，为了能够更清楚了解本发明的技术手段，而可依照说明书的内容予以实施，并且为了让本发明的上述和其它目的、特征和优点能够更明显易懂，以下特举本发明的具体实施方式。The above description is only an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the description, and in order to make the above and other purposes, features and advantages of the present invention more obvious and understandable , the specific embodiments of the present invention are enumerated below.
附图说明Description of drawings
通过阅读下文优选实施方式的详细描述，各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。附图仅用于示出优选实施方式的目的，而并不认为是对本发明的限制。而且在整个附图中，用相同的参考符号表示相同的部件。在附图中：Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiment. The drawings are only for the purpose of illustrating a preferred embodiment and are not to be considered as limiting the invention. Also throughout the drawings, the same reference numerals are used to designate the same parts. In the attached picture:
图1是现有技术的视频推荐方法的流程图；Fig. 1 is a flowchart of a video recommendation method in the prior art;
图2是根据本发明的实施例的基于历史信息的视频推荐方法的流程图；Fig. 2 is the flowchart of the video recommendation method based on historical information according to an embodiment of the present invention;
图3是根据本发明的实施例的基于历史信息的视频推荐装置的框图；以及3 is a block diagram of a video recommendation device based on historical information according to an embodiment of the present invention; and
图4是根据本发明的实施例的视频推荐模块的框图。FIG. 4 is a block diagram of a video recommendation module according to an embodiment of the present invention.
具体实施方式Detailed ways
下面将参照附图更详细地描述本公开的示例性实施例。虽然附图中显示了本公开的示例性实施例，然而应当理解，可以以各种形式实现本公开而不应被这里阐述的实施例所限制。相反，提供这些实施例是为了能够更透彻地理解本公开，并且能够将本公开的范围完整的传达给本领域的技术人员。Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided for more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.
图2示意性地图示了根据本发明的实施例的基于历史信息的视频推荐方法的流程图。如图2所示，在根据本发明的实施例的视频推荐方法100中，一开始，执行步骤S101：获取用户的视频观看记录信息；接着，执行步骤S103：根据所述视频观看记录信息，计算用户观看过的各类型视频的类型需求强度；然后，执行步骤S105：对于每一类型视频，根据视频特征进行分组，并且获取各特征组的特征需求强度；最后，执行步骤S107：基于所述类型需求强度和/或所述特征需求强度，向用户推荐视频。下面对上面各步骤进行详细描述。Fig. 2 schematically illustrates a flowchart of a video recommendation method based on historical information according to an embodiment of the present invention. As shown in Figure 2, in the video recommendation method 100 according to the embodiment of the present invention, at the beginning, step S101 is executed: obtain the user's video viewing record information; then, step S103 is executed: according to the video viewing record information, calculate The type demand strength of each type of video that the user has watched; then, perform step S105: for each type of video, group according to the video feature, and obtain the feature demand strength of each feature group; finally, perform step S107: based on the type The strength of demand and/or the strength of demand for the feature recommends videos to the user. The above steps are described in detail below.
首先，在步骤S101中，获取用户的视频观看记录信息。根据本发明的实施例，用户的Cookie文件可以包含所述视频观看记录信息，即，在用户每次观看在线视频时，都会在Cookie文件中留下视频观看记录，该视频观看记录可以至少包括用户所观看的所有视频的ID名称、类型、特征、观看时间、作者、片长、发行者等信息。然而，本发明的范围并不限于此，视频观看记录信息还可以包含在用户端或者服务器端的其它文件中。First, in step S101, the user's video watching record information is acquired. According to an embodiment of the present invention, the user's cookie file may contain the video viewing record information, that is, each time the user watches an online video, a video viewing record will be left in the cookie file, and the video viewing record may at least include the user's ID name, type, feature, viewing time, author, length, publisher and other information of all videos watched. However, the scope of the present invention is not limited thereto, and the video viewing record information may also be included in other files on the client side or server side.
在用户通过例如浏览器访问在线视频网站时，浏览器会向网站服务器发送页面请求，其中包括Cookie文件，此时，就可以得到用户的Cookie文件，并获取其中的用户的视频观看记录信息。When a user accesses an online video website through, for example, a browser, the browser will send a page request to the website server, including a cookie file. At this time, the user's cookie file and the user's video viewing record information can be obtained.
接下来，在步骤S103中，根据所述视频观看记录信息，计算用户观看过的各类型视频的类型需求强度。根据本发明的实施例，所述步骤S103可以包括子步骤S103a、S103b、S103c、以及S103d。Next, in step S103, according to the video viewing record information, the type demand intensity of each type of video watched by the user is calculated. According to an embodiment of the present invention, the step S103 may include sub-steps S103a, S103b, S103c, and S103d.
在子步骤S103a中，统计用户观看过的视频的类型，此处可以使用count distinct函数，来计算非重复类型的总数量TypeN。例如，通过Cookie文件中的用户视频观看记录信息，发现用户观看过3个电影、2个电视剧、1个动漫剧，按观看时间新旧，电影最新，电视剧次之，动漫剧最旧，类型总数量TypeN为3。In sub-step S103a, the types of videos watched by the user are counted, and the count distinct function can be used here to calculate the total number TypeN of the distinct types. For example, through the user's video viewing record information in the cookie file, it is found that the user has watched 3 movies, 2 TV dramas, and 1 anime drama. According to the viewing time, the movie is the newest, followed by the TV series, and the animation drama is the oldest. The total number of genres TypeN is 3.
接着，在子步骤S103b中，对于每一类型视频，根据该类型视频数量和所有视频总数量，计算该类型视频的内容需求强度，其计算公式为：ContentReqi＝视频中类型为Typei的视频的数量÷所有视频总数量。仍以上面的情况为例，三种类型的视频的内容需求强度分别是，电影：3，电视剧：2，动漫剧：1。Then, in sub-step S103b, for each type of video, according to the number of videos of this type and the total number of all videos, the content demand strength of this type of video is calculated, and its calculation formula is: ContentReq i =type in the video is the video of Type i of ÷ the total number of all videos. Still taking the above situation as an example, the content demand strengths of the three types of videos are respectively, movies: 3, TV dramas: 2, and animation dramas: 1.
然后，在子步骤S103c中，根据该类型视频的观看时间在所有视频的观看时间中所处的时间先后位置，计算该类型视频的时间需求强度，其计算公式为：FreshReqi＝Typei视频的观看时间的时间先后位置÷所有类型视频的观看时间的时间先后位置之和。仍以上面的情况为例，三种类型的视频的时间需求强度分别是，电影：3，电视剧：2，动漫剧：1。可选地，只取每种类型中观看时间最新的一个视频的观看时间进行计算。Then, in sub-step S103c, calculate the time demand strength of this type of video according to the viewing time of this type of video in the chronological position of all video viewing times, its calculation formula is: FreshReq i =Type i video The chronological position of the viewing time ÷ the sum of the chronological positions of the viewing time of all types of videos. Still taking the above situation as an example, the time demand strengths of the three types of videos are respectively, movies: 3, TV dramas: 2, and animation dramas: 1. Optionally, only the viewing time of a video with the latest viewing time in each type is used for calculation.
最后，在子步骤S103d中，可以基于所述内容需求强度和所述时间需求强度，计算该类型视频的类型需求强度，其计算公式为：Finally, in sub-step S103d, based on the content demand strength and the time demand strength, the type demand strength of this type of video can be calculated, and the calculation formula is:
TypeReqi＝a*ContentReqi+(1-a)*FreshReqi TypeReq i ＝a*ContentReq i +(1-a)*FreshReq i
其中，a是拟合参数，可以根据实际需要进行选取，例如取a＝0.5。在上面的例子中，三种类型的视频的类型需求强度分别是，电影：3，电视剧：2，动漫剧：1。Wherein, a is a fitting parameter, which can be selected according to actual needs, for example, a=0.5. In the above example, the genre demand strengths of the three types of videos are respectively, movies: 3, TV dramas: 2, and anime dramas: 1.
在步骤S103之后，执行步骤S105：对于每一类型视频，根据视频特征进行分组，并且获取各特征组的特征需求强度。其中，所述对于每一类型视频、根据视频特征进行分组的步骤包括子步骤S105a和S105b。After step S103, step S105 is executed: for each type of video, group according to video features, and obtain the feature demand strength of each feature group. Wherein, the step of grouping according to video features for each type of video includes sub-steps S105a and S105b.
在子步骤S105a中，根据该类型视频中各视频的特征，确定该类型视频中的各特征组，可以利用count distinct函数来确定非重复特征组的数量。可选地，子步骤S105a可以利用Canopy聚类算法执行。在Canopy聚类算法中，设置第一距离阈值和第二距离阈值，其中所述第一距离阈值小于所述第二距离阈值，这两个距离阈值的值可以根据实际需要来选取；将特征差异小于所述第一距离阈值的视频分入相同的特征组中；将与一特征组的特征差异小于所述第二距离阈值、但大于所述第一距离阈值的视频分入该特征组，并且另外分入单独的特征组中；根据特征组中的视频，计算各特征组的中心特征。因为每个特征组中视频间的差异都很小(相似度很高)，所以Canopy聚类算法能够给出特征组数和各组中心特征；因为某视频可能会被分到所有与其特征差异小于所述第二特征阈值的特征组中，即一个视频会分到多个特征组中，因此这是一个有交叉的分组方式。In sub-step S105a, according to the characteristics of each video in this type of video, each feature group in this type of video is determined, and the count distinct function can be used to determine the number of non-repeating feature groups. Optionally, sub-step S105a can be performed using the Canopy clustering algorithm. In the Canopy clustering algorithm, a first distance threshold and a second distance threshold are set, wherein the first distance threshold is smaller than the second distance threshold, and the values of these two distance thresholds can be selected according to actual needs; the feature difference Classifying videos less than the first distance threshold into the same feature group; classifying videos with feature differences from a feature group that are smaller than the second distance threshold but greater than the first distance threshold into the feature group, and In addition, it is divided into a separate feature group; according to the video in the feature group, the central feature of each feature group is calculated. Because the difference between videos in each feature group is very small (high similarity), the Canopy clustering algorithm can give the number of feature groups and the center features of each group; because a video may be divided into all its feature differences less than In the feature group of the second feature threshold, that is, one video will be divided into multiple feature groups, so this is a grouping method with crossover.
然后，在子步骤S105b中，将该类型视频中的各视频根据其特征分入各特征组中。可选地，子步骤S105b可以利用K-Means聚类算法来执行，K-Means聚类算法的计算公式为：Clusteri＝[视频j组成的集合，如果视频j距离第i个聚类中心点最近的话]。在K-Means聚类算法中，初始设定K为Canopy聚类的组数，K个中心点为Canopy聚类分组的中心点，循环聚类直至K个中心点基本不移动(新中心点与原中心点距离足够小)；每轮聚类是计算所有视频与K个中心点的距离，将其分到距离最小的中心点对应的组中，所有视频分组完成后，根据每个组中的全部视频，重新计算K个中心点。因为此算法收敛速度较快，因此K-Means聚类是一个高效算法；而算法收敛时，再进行一轮计算，分组不会优化，因此该算法还是一个准确率高的算法；另外，所有视频均只分到了一个距离最近的组中，因此这是一个无交叉的分组方式。Then, in sub-step S105b, each video in this type of video is classified into each feature group according to its feature. Optionally, sub-step S105b can be performed using the K-Means clustering algorithm, and the calculation formula of the K-Means clustering algorithm is: Cluster i =[a set composed of video j , if video j is far from the i-th cluster center point recent words]. In the K-Means clustering algorithm, K is initially set as the number of Canopy clustering groups, and K center points are the center points of Canopy clustering groups, and the clustering is repeated until K center points basically do not move (new center points and The original center point distance is small enough); each round of clustering is to calculate the distance between all videos and K center points, and divide them into the group corresponding to the center point with the smallest distance. After all videos are grouped, according to each group For all videos, recalculate K center points. Because this algorithm converges faster, K-Means clustering is an efficient algorithm; when the algorithm converges, another round of calculation is performed, and the grouping will not be optimized, so this algorithm is still a high accuracy algorithm; in addition, all videos are only assigned to the closest group, so this is a non-crossover grouping.
举例而言，假设用户观看了“无间道1”、“无间道2”、“桃姐”这3部电影，“战火西北狼”、“童话二分之一”这2部电视剧，“海贼王”这1部动漫剧。对电影类型进行聚类，聚类数为2，第一特征组包括“无间道1”和“无间道2”，第二特征组包括“桃姐”。对电视剧类型进行聚类，聚类数为2，第一特征组包括“战火西北狼”，第二特征组包括“童话二分之一”，动漫聚类特征组为1个，包括“海贼王”。分别对每个特征组的视频进行推荐，电影类型的第一特征组可以推荐出“无间道3”、“窃听风云”等警匪电影，电影类型的第二特征组可以推荐出“天水围的雾与夜”、“海洋天堂”等情感电影；电视剧类型的第一特征组可以推荐出“正者无敌”、“尖刀队”等军事剧，电视剧类型的第二特征组可以推荐出“爱情公寓3”、“我家有喜”等爱情剧；动漫剧类型的特征组可以推荐出“海贼王娜美篇”等冒险动漫剧。For example, assume that the user has watched 3 movies "Infernal Affairs 1", "Infernal Affairs 2" and "Sister Tao", 2 TV series "War in the Northwest Wolf" and "Half of a Fairy Tale", "One Piece" This is an anime drama. The movie genre is clustered, the number of clusters is 2, the first feature group includes "Infernal Affairs 1" and "Infernal Affairs 2", and the second feature group includes "Peach Sister". Perform clustering on TV drama types, the number of clusters is 2, the first feature group includes "Northwest Wolf", the second feature group includes "fairy tale half", and the animation cluster feature group is 1, including "One Piece" . The videos of each feature group are recommended separately. The first feature group of the movie type can recommend police and gangster movies such as "Infernal Affairs 3" and "Eavesdropping". The second feature group of the movie type can recommend "The Fog in Tianshuiwei". Emotional movies such as "Yu Ye" and "Ocean Paradise"; the first feature group of TV series can recommend military dramas such as "Invincible", "Sharp Knife Team", and the second feature group of TV series can recommend "Love Apartment 3 ", "My Family Has Love" and other romantic dramas; the feature group of animation drama types can recommend adventure animation dramas such as "One Piece Nami".
根据本发明的实施例，特征组的特征需求强度可以是根据该特征组中的视频的观看时间在所有特征组的视频的观看时间中所处的时间先后位置确定的。可选地，所述特征需求强度可以是根据该特征组中最新观看的视频的观看时间在各特征组中最新观看的视频的观看时间中所处的时间先后位置确定的。例如，在上面的例子中，假设在电影类型的视频中，第一特征组中的“无间道2”是该特征组中最新观看的视频，而第二特征组中的“桃姐”是该特征组中最新观看的视频，而“无间道2”的观看时间又比“桃姐”的观看时间新。因此，在电影类型中，第一特征组的特征需求强度高于第二特征组的特征需求强度。According to an embodiment of the present invention, the feature demand intensity of a feature group may be determined according to the chronological positions of the viewing time of videos in the feature group among the viewing times of videos in all feature groups. Optionally, the feature demand intensity may be determined according to the chronological position of the viewing time of the latest watched video in the feature group among the viewing times of the latest watched video in each feature group. For example, in the above example, suppose that among the movie-type videos, "Infernal Affairs 2" in the first feature group is the most recently watched video in the feature group, and "Peach Sister" in the second feature group is the The most recently watched video in the feature group, and the viewing time of "Infernal Affairs 2" is newer than that of "Sister Tao". Therefore, in the movie genre, the characteristic demand intensity of the first characteristic group is higher than that of the second characteristic group.
最后，在步骤S107中，基于所述类型需求强度和/或所述特征需求强度，向用户推荐视频。根据本发明的实施例，步骤S107可以包括子步骤S107a和/或S107b。Finally, in step S107, videos are recommended to the user based on the genre demand strength and/or the feature demand strength. According to an embodiment of the present invention, step S107 may include sub-steps S107a and/or S107b.
在子步骤S107a中，按照类型需求强度从高到低的顺序，向用户推荐各类型的视频。以上面的情况(时间需求强度，电影：3，电视剧：2，动漫剧：1)为例，首先向用户推荐电影，其次向用户推荐电视剧，最后向用户推荐动漫剧。例如，可以在界面中展现“电影推荐”、“电视剧推荐”、“动漫剧推荐”3个链接按钮，默认展现“电影推荐”的内容。另外，可以响应于用户更换视频类型的请求，切换向用户推荐的视频类型，例如，用户可以通过选择“电视剧推荐”或“动漫剧推荐”来分别观看电视剧类型视频和动漫剧类型视频中的推荐视频。In sub-step S107a, various types of videos are recommended to the user in descending order of type demand intensity. Taking the above situation (time demand intensity, movies: 3, TV dramas: 2, anime dramas: 1) as an example, movies are recommended to users first, TV dramas are recommended to users second, and anime dramas are recommended to users last. For example, three link buttons "recommended movies", "recommended TV dramas" and "recommended cartoons" can be displayed in the interface, and the content of "recommended movies" is displayed by default. In addition, in response to the user's request to change the video type, the type of video recommended to the user can be switched. For example, the user can watch the recommendations in the TV series type video and animation type video by selecting "TV series recommendation" or "Anime series recommendation" respectively. video.
在子步骤S107b中，对于每一类型视频，按照特征需求强度从高到低的顺序，向用户推荐各特征组的视频。例如，在上面的例子中，在“电影推荐”中，可以在左边展现电影类型中特征需求强度最高的第一特征组中之前观看过的电影“无间道1”、“无间道2”，右边展现第一特征组中的推荐电影“无间道3”、“窃听风云”等。而如果用户选择了“电视剧推荐”，则可以向用户展现电视剧类型中特征需求强度最高的第一特征组中的推荐视频，例如，左边展现第一特征组中已经观看过的观看电视剧“战火西北狼”，右边展现第一特征组中推荐的电视剧“正者无敌”、“尖刀队”等。In sub-step S107b, for each type of video, videos of each feature group are recommended to the user in order of feature demand intensity from high to low. For example, in the above example, in "Movie Recommendation", the previously watched movies "Infernal Affairs 1" and "Infernal Affairs 2" in the first feature group with the highest feature demand intensity in the movie genre can be displayed on the left, and the movies "Infernal Affairs 2" on the right The recommended movies "Infernal Affairs 3", "Eavesdropping" and the like in the first feature group are displayed. And if the user selects "TV drama recommendation", the recommended video in the first feature group with the highest feature demand intensity in the TV drama type can be shown to the user. "Wolf", on the right is the recommended TV series "Invincible", "Sharp Knife Team" and so on in the first feature group.
另外，还可以响应于用户更换视频特征组的请求，切换向用户推荐的视频特征组。即，向用户提供切换到其它特征组视频推荐的方式，例如，在界面中提供“换一换”链接按钮。如上例，在“电影推荐”类型下，通过点击“换一换”链接按钮，可以由第一特征组电影推荐切换到第二特征组电影推荐：左边展现第二特征组中观看过的电影“桃姐”，右边展现第二特征组中推荐的“天水围的雾与夜”、“海洋天堂”等情感电影；通过再次点击“换一换”链接按钮，可以切换回第一组电影推荐。In addition, in response to the user's request to change the video feature set, the video feature set recommended to the user may be switched. That is, provide the user with a way to switch to video recommendations of other feature groups, for example, provide a "Change" link button in the interface. As in the above example, under the "Movie Recommendation" type, by clicking the "Change" link button, you can switch from the movie recommendation of the first feature group to the movie recommendation of the second feature group: the movies watched in the second feature group are displayed on the left " Peach Sister”, on the right side are emotional movies such as “Fog and Night in Tianshuiwei” and “Ocean Paradise” recommended in the second feature group; by clicking the “Change” link button again, you can switch back to the first group of movie recommendations.
本领域技术人员可以容易理解，上述界面展示的方式仅为示例，用于帮助读者理解本发明的原理，而非将本发明的范围限制于此，还可以采用其它各种方式安排界面，以向用户推荐视频。Those skilled in the art can easily understand that the display of the above-mentioned interface is only an example to help readers understand the principle of the present invention, rather than limit the scope of the present invention, and the interface can also be arranged in various other ways to provide User recommended videos.
本发明提供了一种视频推荐方法。根据本发明的实施例，根据用户的视频观看记录信息，计算各类型视频的类型需求强度，将每一类型视频中的视频分入各特征组，并获取各特征组的特征需求强度，最后基于类型需求强度和/或特征需求强度向用户推荐视频，可以简单有效地提供多类型和多特征的视频推荐结果，满足了用户复杂的视频推荐需求。相对于传统的视频推荐方法，本发明能够计算用户对于不同类型视频的类型需求强度，并据此以不同的优先级推荐电影、电视剧、动漫剧、综艺节目、体育节目等各种类型的视频，并且根据用户的特征需求强度，以不同的优先级推荐特定视频类型下的不同特征的视频，如电影类型中的爱情片、科幻片、战争片，电视剧类型中家庭剧、历史剧等等，这样，可以分组对用户进行视频推荐，并且优先提供用户最喜欢的特征组的视频。另外，还可以响应于用户的请求，切换向用户推荐的视频类型和/或视频特征组。即，用户可以通过点击其他类型的视频而切换到该类型的视频推荐结果；用户也可以通过点击“换一换”按钮，切换到其他特征所对应的视频推荐结果。The invention provides a video recommendation method. According to an embodiment of the present invention, according to the user's video viewing record information, the type demand strength of each type of video is calculated, the videos in each type of video are divided into each feature group, and the feature demand strength of each feature group is obtained, and finally based on The genre demand strength and/or feature demand strength recommend videos to users, which can simply and effectively provide multi-type and multi-feature video recommendation results, and meet the complex video recommendation needs of users. Compared with the traditional video recommendation method, the present invention can calculate the type demand intensity of users for different types of videos, and accordingly recommend various types of videos such as movies, TV dramas, animation dramas, variety shows, sports programs, etc. with different priorities, And according to the strength of the user's characteristic needs, videos with different characteristics under a specific video type are recommended with different priorities, such as romance films, science fiction films, and war films in movie types, family dramas, historical dramas, etc. in TV series, etc. In this way, Video recommendations can be made to users in groups, and videos of the user's favorite feature group are given priority. In addition, the video type and/or video feature group recommended to the user may also be switched in response to the user's request. That is, the user can switch to video recommendation results of other types by clicking on other types of videos; the user can also switch to video recommendation results corresponding to other features by clicking the "Change" button.
与上述的方法100相对应，本发明还提供了一种基于历史信息的视频推荐装置200。图3示意性地图示了根据本发明的实施例的基于历史信息的视频推荐装置200的框图，参见图3，该装置200包括：Corresponding to the above method 100, the present invention also provides an apparatus 200 for recommending videos based on historical information. FIG. 3 schematically illustrates a block diagram of a video recommendation device 200 based on historical information according to an embodiment of the present invention. Referring to FIG. 3 , the device 200 includes:
信息获取模块201，用于获取用户的视频观看记录信息，该信息获取模块201可以用于执行方法100中的步骤S101；The information obtaining module 201 is used to obtain the user's video viewing record information, and the information obtaining module 201 can be used to execute step S101 in the method 100;
类型需求强度计算模块203，用于根据所述视频观看记录信息，计算用户观看过的各类型视频的类型需求强度，该类型需求强度计算模块203可以用于执行方法100中的步骤S103；The type demand strength calculation module 203 is used to calculate the type demand strength of each type of video watched by the user according to the video viewing record information, and the type demand strength calculation module 203 can be used to perform step S103 in the method 100;
特征需求强度获取模块205，用于对于每一类型视频，根据视频特征进行分组，并且获取各特征组的特征需求强度，该特征需求强度获取模块205可以用于执行方法100中的步骤S105；以及The feature demand strength acquisition module 205 is used to group each type of video according to video features, and acquire the feature demand strength of each feature group, and the feature demand strength acquisition module 205 can be used to perform step S105 in the method 100; and
视频推荐模块207，用于基于所述类型需求强度和/或所述特征需求强度，向用户推荐视频，该视频推荐模块207可以用于执行方法100中的步骤S107。The video recommending module 207 is configured to recommend videos to the user based on the genre demand strength and/or the feature demand strength, and the video recommending module 207 can be used to execute step S107 in the method 100 .
在本发明的实施例中，所述类型需求强度计算模块203包括：In an embodiment of the present invention, the type demand intensity calculation module 203 includes:
类型统计子模块203a，用于统计用户观看过的视频的类型，其可以用于执行方法100的步骤S103中的子步骤S103a；The type statistics submodule 203a is used to count the types of videos watched by the user, which can be used to perform the substep S103a in the step S103 of the method 100;
内容需求强度计算子模块203b，用于对于每一类型视频，根据该类型视频数量和所有视频总数量，计算该类型视频的内容需求强度，其可以用于执行方法100的步骤S103中的子步骤S103b；The content demand intensity calculation sub-module 203b is used to calculate the content demand intensity of this type of video according to the number of videos of this type and the total number of all videos for each type of video, which can be used to execute the sub-steps in step S103 of method 100 S103b;
时间需求强度计算子模块203c，用于根据该类型视频的观看时间在所有视频的观看时间中所处的时间先后位置，计算该类型视频的时间需求强度，其可以用于执行方法100的步骤S103中的子步骤S103c；以及The time demand intensity calculation sub-module 203c is used to calculate the time demand intensity of this type of video according to the chronological position of the viewing time of this type of video in the viewing time of all videos, which can be used to execute step S103 of the method 100 Substep S103c in; and
类型需求强度计算子模块203d，用于基于所述内容需求强度和所述时间需求强度，计算该类型视频的类型需求强度，其可以用于执行方法100的步骤S103中的子步骤S103d，其可以基于上面子步骤S103d中的公式来计算所述类型需求强度。The type demand strength calculation sub-module 203d is used to calculate the type demand strength of this type of video based on the content demand strength and the time demand strength, which can be used to execute the sub-step S103d in the step S103 of the method 100, which can The category demand intensity is calculated based on the formula in the above sub-step S103d.
在本发明的实施例中，所述特征需求强度获取模块205包括：In an embodiment of the present invention, the feature demand intensity acquisition module 205 includes:
特征组确定子模块205a，用于根据该类型视频中各视频的特征，确定该类型视频中的各特征组，其可以用于执行方法100的步骤S105中的子步骤S105a；以及The feature group determination submodule 205a is used to determine each feature group in this type of video according to the characteristics of each video in this type of video, which can be used to perform substep S105a in step S105 of method 100; and
特征分组子模块205b，用于将该类型视频中的各视频根据其特征分入各特征组中，其可以用于执行方法100的步骤S105中的子步骤S105b。The feature grouping sub-module 205b is configured to classify each video of the type of video into each feature group according to its feature, which can be used to execute sub-step S105b in step S105 of method 100 .
在本发明的实施例中，所述特征组确定子模块205a利用Canopy聚类算法，根据该类型视频中各视频的特征，确定该类型视频中的各特征组，其中所述特征组确定子模块205a：In an embodiment of the present invention, the feature group determining submodule 205a utilizes the Canopy clustering algorithm to determine each feature group in this type of video according to the characteristics of each video in this type of video, wherein the feature group determining submodule 205a:
设置第一距离阈值和第二距离阈值，其中所述第一距离阈值小于所述第二距离阈值；setting a first distance threshold and a second distance threshold, wherein the first distance threshold is smaller than the second distance threshold;
将特征差异小于所述第一距离阈值的视频分入相同的特征组中；Classifying videos with feature differences smaller than the first distance threshold into the same feature group;
将与一特征组的特征差异小于所述第二距离阈值、但大于所述第一距离阈值的视频分入该特征组，并且另外分入单独的特征组中；以及sorting videos whose features differ from a feature group by less than said second distance threshold but greater than said first distance threshold into that feature group, and additionally into a separate feature group; and
根据特征组中的视频，计算各特征组的中心特征。Based on the videos in the feature group, calculate the central feature of each feature group.
在本发明的实施例中，所述特征分组子模块205b利用K-Means聚类算法将该类型视频中的各视频根据其特征分入各特征组中，其中所述特征分组子模块205b：In an embodiment of the present invention, the feature grouping submodule 205b uses the K-Means clustering algorithm to classify each video in this type of video into each feature group according to its characteristics, wherein the feature grouping submodule 205b:
计算所述各视频与各特征组的中心特征的差异；Calculate the difference between each video and the central feature of each feature group;
将所述各视频分入与其差异最小的特征组中；Classify each video into the feature group with the smallest difference;
根据特征组中的视频，重新计算各特征组的中心特征；以及recalculate the central feature of each feature group based on the videos in the feature group; and
重复执行上述步骤，直到所述各特征组的中心特征与前一次计算的各特征组的中心特征之间的差异小于预先定义的阈值为止。The above steps are repeatedly executed until the difference between the center feature of each feature group and the center feature of each feature group calculated last time is smaller than a predefined threshold.
在本发明的实施例中，所述特征需求强度获取模块205根据该特征组中的视频的观看时间在所有特征组的视频的观看时间中所处的时间先后位置，确定特征组的特征需求强度。In an embodiment of the present invention, the feature demand strength acquisition module 205 determines the feature demand strength of the feature group according to the chronological positions of the video viewing time in the feature group among the video viewing times of all feature groups .
在本发明的实施例中，所述特征需求强度获取模块205根据该特征组中最新观看的视频的观看时间在各特征组中最新观看的视频的观看时间中所处的时间先后位置，确定特征组的特征需求强度。In an embodiment of the present invention, the feature demand strength acquisition module 205 determines the feature according to the time sequence position of the viewing time of the latest video in the feature group in the viewing time of the latest video in each feature group. The characteristic demand strength of the group.
在本发明的实施例中，参见图4，所述视频推荐模块207可以包括：In an embodiment of the present invention, referring to FIG. 4, the video recommendation module 207 may include:
类型推荐子模块207a，用于按照类型需求强度从高到低的顺序，向用户推荐各类型的视频，其可以用于执行方法100的步骤S107中的子步骤S107a；以及The type recommendation sub-module 207a is used to recommend various types of videos to the user in order of type demand intensity from high to low, which can be used to perform sub-step S107a in step S107 of method 100; and
特征组推荐子模块207b，用于对于每一类型视频，按照特征需求强度从高到低的顺序，向用户推荐各特征组的视频，其可以用于执行方法100的步骤S107中的子步骤S107b。The feature group recommending sub-module 207b is used to recommend videos of each feature group to the user in order of feature demand strength from high to low for each type of video, which can be used to execute sub-step S107b in step S107 of method 100 .
在本发明的实施例中，所述视频推荐模块207还可以包括：In an embodiment of the present invention, the video recommendation module 207 may also include:
类型切换子模块，用于响应于用户更换视频类型的请求，切换向用户推荐的视频类型；和/或A type switching submodule, configured to switch the type of video recommended to the user in response to the user's request for changing the type of video; and/or
特征组切换子模块，用于响应于用户更换视频特征组的请求，切换向用户推荐的视频特征组。The feature group switching sub-module is used to switch the video feature group recommended to the user in response to the user's request to change the video feature group.
在本发明的实施例中，所述视频观看记录信息包含在用户的Cookie文件中。In an embodiment of the present invention, the video watching record information is included in the user's cookie file.
由于上述各装置实施例与前述各方法实施例相对应，因此不再对各装置实施例进行详细描述。Since the above-mentioned device embodiments correspond to the above-mentioned method embodiments, no detailed description will be given for each device embodiment.
在此提供的算法和显示不与任何特定计算机、虚拟系统或者其它设备固有相关。各种通用系统也可以与基于在此的示教一起使用。根据上面的描述，构造这类系统所要求的结构是显而易见的。此外，本发明也不针对任何特定编程语言。应当明白，可以利用各种编程语言实现在此描述的本发明的内容，并且上面对特定语言所做的描述是为了披露本发明的最佳实施方式。The algorithms and displays presented herein are not inherently related to any particular computer, virtual system, or other device. Various generic systems can also be used with the teachings based on this. The structure required to construct such a system is apparent from the above description. Furthermore, the present invention is not specific to any particular programming language. It should be understood that various programming languages can be used to implement the content of the present invention described herein, and the above description of specific languages is for disclosing the best mode of the present invention.
在此处所提供的说明书中，说明了大量具体细节。然而，能够理解，本发明的实施例可以在没有这些具体细节的情况下实践。在一些实例中，并未详细示出公知的方法、结构和技术，以便不模糊对本说明书的理解。In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure the understanding of this description.
类似地，应当理解，为了精简本公开并帮助理解各个发明方面中的一个或多个，在上面对本发明的示例性实施例的描述中，本发明的各个特征有时被一起分组到单个实施例、图、或者对其的描述中。然而，并不应将该公开的方法解释成反映如下意图：即所要求保护的本发明要求比在每个权利要求中所明确记载的特征更多的特征。更确切地说，如下面的权利要求书所反映的那样，发明方面在于少于前面公开的单个实施例的所有特征。因此，遵循具体实施方式的权利要求书由此明确地并入该具体实施方式，其中每个权利要求本身都作为本发明的单独实施例。Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, in order to streamline this disclosure and to facilitate an understanding of one or more of the various inventive aspects, various features of the invention are sometimes grouped together in a single embodiment, figure, or its description. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the Detailed Description are hereby expressly incorporated into this Detailed Description, with each claim standing on its own as a separate embodiment of this invention.
本领域那些技术人员可以理解，可以对实施例中的装置中的模块进行自适应性地改变并且把它们设置在与该实施例不同的一个或多个装置中。可以把实施例中的若干模块组合成一个模块或单元或组件，以及此外可以把它们分成多个子模块或子单元或子组件。除了这样的特征和/或过程或者模块中的至少一些是相互排斥之外，可以采用任何组合对本说明书(包括伴随的权利要求、摘要和附图)中公开的所有特征以及如此公开的任何方法或者设备的所有过程或单元进行组合。除非另外明确陈述，本说明书(包括伴随的权利要求、摘要和附图)中公开的每个特征可以由提供相同、等同或相似目的替代特征来代替。Those skilled in the art can understand that the modules in the device in the embodiment can be adaptively changed and arranged in one or more devices different from the embodiment. Several modules in an embodiment can be combined into one module or unit or assembly, and further they can be divided into a plurality of sub-modules or sub-units or sub-assemblies. All features disclosed in this specification (including accompanying claims, abstract and drawings) and any method or method so disclosed may be used in any combination, except that at least some of such features and/or procedures or modules are mutually exclusive. All processes or units of equipment are combined. Each feature disclosed in this specification (including accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
此外，本领域的技术人员能够理解，尽管在此所述的一些实施例包括其它实施例中所包括的某些特征而不是其它特征，但是不同实施例的特征的组合意味着处于本发明的范围之内并且形成不同的实施例。例如，在权利要求书中，所要求保护的实施例的任意之一都可以以任意的组合方式来使用。Furthermore, those skilled in the art will understand that although some embodiments described herein include some features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention. and form different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.
本发明的各个装置实施例可以以硬件实现，或者以在一个或者多个处理器上运行的软件模块实现，或者以它们的组合实现。本领域的技术人员应当理解，可以在实践中使用微处理器或者数字信号处理器(DSP)来实现根据本发明实施例的装置中的一些或者全部模块的一些或者全部功能。本发明还可以实现为用于执行这里所描述的方法的一部分或者全部的装置程序(例如，计算机程序和计算机程序产品)。这样的实现本发明的程序可以存储在计算机可读介质上，或者可以具有一个或者多个信号的形式。这样的信号可以从因特网网站上下载得到，或者在载体信号上提供，或者以任何其他形式提供。The various apparatus embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) may be used in practice to implement some or all functions of some or all modules in the device according to the embodiment of the present invention. The present invention can also be implemented as an apparatus program (for example, a computer program and a computer program product) for performing a part or all of the methods described herein. Such a program for realizing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet site, or provided on a carrier signal, or provided in any other form.
应该注意的是上述实施例对本发明进行说明而不是对本发明进行限制，并且本领域技术人员在不脱离所附权利要求的范围的情况下可设计出替换实施例。在权利要求中，不应将位于括号之间的任何参考符号构造成对权利要求的限制。单词“包含”不排除存在未列在权利要求中的元件或步骤。位于元件之前的单词“一”或“一个”不排除存在多个这样的元件。本发明可以借助于包括有若干不同元件的硬件以及借助于适当编程的计算机来实现。在列举了若干装置的单元权利要求中，这些装置中的若干个可以是通过同一个硬件项来具体体现。单词第一、第二、以及第三等的使用不表示任何顺序。可将这些单词解释为名称。It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means can be embodied by one and the same item of hardware. The use of the words first, second, and third, etc. does not indicate any order. These words can be interpreted as names.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510341831.XA CN104918118B (en) | 2012-10-24 | 2012-10-24 | Video recommendation method and device based on historical information |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210408686.9A CN102946566B (en) | 2012-10-24 | 2012-10-24 | Video recommendation method and device based on historical information |
| CN201510341831.XA CN104918118B (en) | 2012-10-24 | 2012-10-24 | Video recommendation method and device based on historical information |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210408686.9A Division CN102946566B (en) | 2012-10-24 | 2012-10-24 | Video recommendation method and device based on historical information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104918118A true CN104918118A (en) | 2015-09-16 |
| CN104918118B CN104918118B (en) | 2019-08-02 |
Family
ID=47729466
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510341831.XA Active CN104918118B (en) | 2012-10-24 | 2012-10-24 | Video recommendation method and device based on historical information |
| CN201210408686.9A Expired - Fee Related CN102946566B (en) | 2012-10-24 | 2012-10-24 | Video recommendation method and device based on historical information |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210408686.9A Expired - Fee Related CN102946566B (en) | 2012-10-24 | 2012-10-24 | Video recommendation method and device based on historical information |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN104918118B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105812834A (en) * | 2016-05-10 | 2016-07-27 | 南京大学 | Video recommendation server, recommendation method and pre-caching method based on cluster information |
| CN105847985A (en) * | 2016-03-30 | 2016-08-10 | 乐视控股（北京）有限公司 | Video recommendation method and device |
| CN109379608A (en) * | 2018-09-13 | 2019-02-22 | 武汉斗鱼网络科技有限公司 | A kind of recommended method and relevant device of direct broadcasting room |
| CN110427556A (en) * | 2019-07-30 | 2019-11-08 | 牡丹江师范学院 | A kind of film recommended method based on literature and art study |
| CN111160254A (en) * | 2019-12-30 | 2020-05-15 | 杭州趣维科技有限公司 | Classification method and system suitable for short video recommendation scattering |
| CN112770181A (en) * | 2021-01-12 | 2021-05-07 | 贵州省广播电视信息网络股份有限公司 | Quick verification system and method for recommended content of family group |
| CN113763027A (en) * | 2021-05-10 | 2021-12-07 | 腾讯科技（深圳）有限公司 | Recommendation information processing method, recommendation information generating method and device |
| CN116010646A (en) * | 2023-01-17 | 2023-04-25 | 北京达佳互联信息技术有限公司 | Video recommendation method, device, electronic device and storage medium |
Families Citing this family (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103440335B (en) * | 2013-09-06 | 2016-11-09 | 北京奇虎科技有限公司 | Video recommendation method and device |
| CN103686236A (en) * | 2013-11-19 | 2014-03-26 | 乐视致新电子科技（天津）有限公司 | Method and system for recommending video resource |
| CN103647800B (en) * | 2013-11-19 | 2017-12-12 | 乐视致新电子科技（天津）有限公司 | Recommend the method and system of application resource |
| CN103686237B (en) * | 2013-11-19 | 2017-03-08 | 乐视致新电子科技（天津）有限公司 | Recommend the method and system of video resource |
| CN103678672B (en) * | 2013-12-25 | 2017-05-24 | 北京中兴通德网络科技有限公司 | Method for recommending information |
| CN103995875A (en) * | 2014-05-23 | 2014-08-20 | 百度在线网络技术（北京）有限公司 | Information recommendation method and device |
| CN105812846B (en) * | 2014-12-30 | 2019-02-01 | Tcl集团股份有限公司 | The acquisition methods and device of network video |
| CN104731950B (en) * | 2015-03-31 | 2018-09-28 | 北京奇艺世纪科技有限公司 | A kind of video recommendation method and device |
| CN104735520A (en) * | 2015-04-01 | 2015-06-24 | 百度在线网络技术（北京）有限公司 | Television program play control method and device and television set |
| CN104853248B (en) * | 2015-05-07 | 2017-09-22 | 海信集团有限公司 | A kind of video recommendation method and device |
| CN104869439B (en) * | 2015-05-14 | 2018-03-09 | 无锡天脉聚源传媒科技有限公司 | A kind of video pushing method and device |
| CN105142025A (en) * | 2015-07-16 | 2015-12-09 | Tcl集团股份有限公司 | Information push method and system based on intelligent television terminal |
| CN105163182B (en) * | 2015-08-24 | 2019-06-11 | Tcl集团股份有限公司 | Method and system for acquiring user behavior of smart TV based on exception mining algorithm |
| CN105554555A (en) * | 2015-12-11 | 2016-05-04 | 广州华多网络科技有限公司 | Channel switching method and device |
| CN105893443A (en) * | 2015-12-15 | 2016-08-24 | 乐视网信息技术（北京）股份有限公司 | Video recommendation method and apparatus, and server |
| CN105915949A (en) * | 2015-12-23 | 2016-08-31 | 乐视网信息技术（北京）股份有限公司 | Video content recommending method, device and system |
| CN105404698A (en) * | 2015-12-31 | 2016-03-16 | 海信集团有限公司 | Education video recommendation method and device |
| CN105678607B (en) * | 2016-01-07 | 2019-05-31 | 合肥工业大学 | A kind of Order Batch method based on improved K-Means algorithm |
| CN105912620A (en) * | 2016-04-07 | 2016-08-31 | 乐视控股（北京）有限公司 | Personalization multimedia recommendation method and device |
| CN105916034A (en) * | 2016-04-14 | 2016-08-31 | 乐视控股（北京）有限公司 | Video recommendation method and apparatus |
| CN105898495A (en) * | 2016-05-26 | 2016-08-24 | 维沃移动通信有限公司 | Method for pushing mobile terminal recommended information and mobile terminal |
| CN107862532B (en) * | 2016-09-22 | 2021-11-26 | 腾讯科技（深圳）有限公司 | User feature extraction method and related device |
| CN106530132A (en) * | 2016-11-14 | 2017-03-22 | 国家电网公司 | Power load clustering method and device |
| CN108205537A (en) * | 2016-12-16 | 2018-06-26 | 北京酷我科技有限公司 | A kind of video recommendation method and system |
| CN107748759A (en) * | 2017-09-21 | 2018-03-02 | 广州华多网络科技有限公司 | Video pushing method and its equipment |
| CN110035302B (en) * | 2018-01-11 | 2021-06-15 | 斑马智行网络(香港)有限公司 | Information recommendation and model training method and device, computing device, storage medium |
| CN108540860B (en) * | 2018-02-28 | 2020-09-01 | 北京奇艺世纪科技有限公司 | Video recall method and device |
| CN111435371B (en) * | 2019-01-15 | 2023-08-11 | 百度在线网络技术（北京）有限公司 | Video recommendation method and system, computer program product and readable storage medium |
| CN111582956A (en) * | 2020-06-29 | 2020-08-25 | 成都新潮传媒集团有限公司 | Advertisement pushing method and device and computer equipment |
| CN112165634B (en) * | 2020-09-29 | 2022-09-16 | 北京百度网讯科技有限公司 | Method for establishing audio classification model and method and device for automatically converting video |
| CN114764457A (en) * | 2021-01-04 | 2022-07-19 | 中国移动通信有限公司研究院 | Video processing method and device, electronic equipment and readable storage medium |
| CN113704543A (en) * | 2021-07-29 | 2021-11-26 | 青岛海尔科技有限公司 | Movie recommendation method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101094372A (en) * | 2007-07-25 | 2007-12-26 | 北京中星微电子有限公司 | Device and method for recommending TV programs |
| CN101287082A (en) * | 2008-05-16 | 2008-10-15 | 华东师范大学 | A Collaborative Filtering Recommendation Method Introducing Program Popularity Weight |
| US20090158343A1 (en) * | 2007-12-07 | 2009-06-18 | Fujitsu Limited | Recording device of broadcasting program |
| CN101483732A (en) * | 2008-01-10 | 2009-07-15 | 北京视博数字电视科技有限公司 | Electronic program guidance generation method and apparatus, program searching method and searching engine for television terminal |
| CN101489107A (en) * | 2009-01-21 | 2009-07-22 | 华东师范大学 | Collaborative filtering recommendation method based on population attribute keyword vector |
| CN102088626A (en) * | 2009-12-02 | 2011-06-08 | Tcl集团股份有限公司 | On-line video recommendation method and video portal service system |
-
2012
- 2012-10-24 CN CN201510341831.XA patent/CN104918118B/en active Active
- 2012-10-24 CN CN201210408686.9A patent/CN102946566B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101094372A (en) * | 2007-07-25 | 2007-12-26 | 北京中星微电子有限公司 | Device and method for recommending TV programs |
| US20090158343A1 (en) * | 2007-12-07 | 2009-06-18 | Fujitsu Limited | Recording device of broadcasting program |
| CN101483732A (en) * | 2008-01-10 | 2009-07-15 | 北京视博数字电视科技有限公司 | Electronic program guidance generation method and apparatus, program searching method and searching engine for television terminal |
| CN101287082A (en) * | 2008-05-16 | 2008-10-15 | 华东师范大学 | A Collaborative Filtering Recommendation Method Introducing Program Popularity Weight |
| CN101489107A (en) * | 2009-01-21 | 2009-07-22 | 华东师范大学 | Collaborative filtering recommendation method based on population attribute keyword vector |
| CN102088626A (en) * | 2009-12-02 | 2011-06-08 | Tcl集团股份有限公司 | On-line video recommendation method and video portal service system |
Non-Patent Citations (2)
| Title |
|---|
| HYEJEONG LEE: "Personalized Recommendation Schemes for DTV Channel Selectors", 《IEEE TRANSACTIONS ON CONSUMER ELECTRONICS》 * |
| 李应安: "《全国优秀硕士学位论文全文数据库 信息科技辑》", 15 February 2012 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105847985A (en) * | 2016-03-30 | 2016-08-10 | 乐视控股（北京）有限公司 | Video recommendation method and device |
| CN105812834A (en) * | 2016-05-10 | 2016-07-27 | 南京大学 | Video recommendation server, recommendation method and pre-caching method based on cluster information |
| CN105812834B (en) * | 2016-05-10 | 2019-03-12 | 南京大学 | Video recommendation server, recommendation method and pre-caching method based on clustering information |
| CN109379608A (en) * | 2018-09-13 | 2019-02-22 | 武汉斗鱼网络科技有限公司 | A kind of recommended method and relevant device of direct broadcasting room |
| CN110427556A (en) * | 2019-07-30 | 2019-11-08 | 牡丹江师范学院 | A kind of film recommended method based on literature and art study |
| CN110427556B (en) * | 2019-07-30 | 2022-10-11 | 牡丹江师范学院 | Film recommendation method based on literary and artistic learning |
| CN111160254A (en) * | 2019-12-30 | 2020-05-15 | 杭州趣维科技有限公司 | Classification method and system suitable for short video recommendation scattering |
| CN112770181A (en) * | 2021-01-12 | 2021-05-07 | 贵州省广播电视信息网络股份有限公司 | Quick verification system and method for recommended content of family group |
| CN113763027A (en) * | 2021-05-10 | 2021-12-07 | 腾讯科技（深圳）有限公司 | Recommendation information processing method, recommendation information generating method and device |
| CN113763027B (en) * | 2021-05-10 | 2023-08-01 | 腾讯科技（深圳）有限公司 | Recommendation information processing method, recommendation information generation method and device |
| CN116010646A (en) * | 2023-01-17 | 2023-04-25 | 北京达佳互联信息技术有限公司 | Video recommendation method, device, electronic device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102946566B (en) | 2015-07-01 |
| CN104918118B (en) | 2019-08-02 |
| CN102946566A (en) | 2013-02-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102946566B (en) | Video recommendation method and device based on historical information | |
| US20230100286A1 (en) | Methods, systems, and media for aggregating and presenting content relevant to a particular video game | |
| US11023100B2 (en) | Methods, systems, and media for creating and updating a group of media content items | |
| US9253511B2 (en) | Systems and methods for performing multi-modal video datastream segmentation | |
| US11055340B2 (en) | System and method for creating synopsis for multimedia content | |
| CN109327714B (en) | Method and system for supplementing live broadcast | |
| US20160014482A1 (en) | Systems and Methods for Generating Video Summary Sequences From One or More Video Segments | |
| CN107657004A (en) | Video recommendation method, system and equipment | |
| CN113111197B (en) | Multimedia content recommendation method, device, equipment and storage medium | |
| US20140304247A1 (en) | Method and System for Content Management | |
| WO2015182064A1 (en) | Information processing device, information processing method, and program | |
| CN111104583A (en) | Live broadcast room recommendation method, storage medium, electronic device and system | |
| US20210377584A1 (en) | Methods, systems, and media for indicating viewership of a video | |
| WO2020026012A1 (en) | Method, apparatus, device/terminal/server and storage medium for content interest collection and content data recommendation | |
| US20250061144A1 (en) | Systems and Methods for Generating Stories for Live Events Using a Scalable Pipeline | |
| US11494450B2 (en) | Providing recommended contents | |
| CN103942247B (en) | Information providing method and device for multimedia resources | |
| CN103593455B (en) | File recommendation method and file recommendation device | |
| CN113742573A (en) | Multimedia resource searching method and device, computer equipment and storage medium | |
| US20150193444A1 (en) | System and method to determine social relevance of Internet content | |
| JP2017073070A (en) | Content recommendation device, content recommendation system, content recommendation method, and program | |
| US9886415B1 (en) | Prioritized data transmission over networks | |
| CN109756759B (en) | A method and device for recommending bullet screen information | |
| CN102968446A (en) | Video recommendation method and video recommendation device | |
| US20160188603A1 (en) | Quotation management platform |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220729 Address after: Room 801, 8th floor, No. 104, floors 1-19, building 2, yard 6, Jiuxianqiao Road, Chaoyang District, Beijing 100015 Patentee after: BEIJING QIHOO TECHNOLOGY Co.,Ltd. Address before: 100088 room 112, block D, 28 new street, new street, Xicheng District, Beijing (Desheng Park) Patentee before: BEIJING QIHOO TECHNOLOGY Co.,Ltd. Patentee before: Qizhi software (Beijing) Co.,Ltd. |
|
| TR01 | Transfer of patent right |