CN107894998A - Video recommendation method and device - Google Patents
Video recommendation method and device Download PDFInfo
- Publication number
- CN107894998A CN107894998A CN201710997564.0A CN201710997564A CN107894998A CN 107894998 A CN107894998 A CN 107894998A CN 201710997564 A CN201710997564 A CN 201710997564A CN 107894998 A CN107894998 A CN 107894998A
- Authority
- CN
- China
- Prior art keywords
- user
- video
- cluster
- clustering
- users
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/71—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/73—Querying
- G06F16/735—Filtering based on additional data, e.g. user or group profiles
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of video recommendation method, collects multiple user behavior datas in the given time, establishes user behavior data sample；Store the user behavior data sample；The video content of the user behavior data sample is clustered using clustering algorithm, obtains the similar user's cluster of at least one interest；Count the video that the similar user's cluster user of the interest was watched；Extract each two field picture of the video and extract feature；The feature of each two field picture of the video is clustered using clustering algorithm, obtains at least one Video clustering；The interactive relation in user's cluster between each user described in the Video clustering is analyzed, the influence power of user's cluster is obtained according to the analysis result；Recommended according to the ranking of the influence power of user's cluster.A kind of video recommendations device is also provided.
Description
Technical Field
The invention relates to the technical field of information, in particular to a video recommendation method and device.
Background
With the rapid development of the Web2.0 technology and the rapid popularization of the online social network, the online social network has become an important carrier for people to maintain social relationships, propagate information and innovate knowledge. The popularization of the mobile internet enables the generation and the release of information to be very convenient and fast, and users can share videos anytime and anywhere. The video platform is not only a content sharing platform, but also a social platform. The user can share videos shot by the user or interested in the videos on the platform, and meanwhile, the user can interact with other users on the platform. Therefore, building a recommendation system by combining the video content and the interaction relationship of the user is a research direction that receives great attention in the industry.
In the prior art, a user actively selects favorite video categories, and then recommends videos of the same category according to records selected by the user; also, other videos of the same category as the video viewed by the user are recommended based on the viewing history of the user. Today, social networks are developed, but the social characteristics of social networks are not fully exploited to predict potential video categories that users may like. Therefore, the existing video recommendation accuracy is not high, and the user experience is poor.
Disclosure of Invention
In view of this, embodiments of the present invention provide a video recommendation method and apparatus, where a part of people with high influence is found, and a suitable video is preferentially recommended to the people, and after some interactions on the video, the people recommend the video to friends of the people. Under the condition of giving the total number of recommendations, more comments or praise of the user on the video can be obtained through the releasing mode, the video is enabled to have a deeper impression, the user experience and interaction frequency are improved, and the user retention rate is improved.
The embodiment of the invention provides a video recommendation method, which comprises the following steps: collecting a plurality of user behavior data within preset time, and establishing a user behavior data sample; storing the user behavior data samples; clustering the video content of the user behavior data sample by adopting a clustering algorithm to obtain at least one user cluster with similar interest; counting videos watched by the user cluster users with similar interests; extracting each frame of image of the video and extracting features; clustering the characteristics of each frame of image of the video by adopting a clustering algorithm to obtain at least one video cluster; analyzing the interactive relation among the users in the user cluster in the video cluster, and obtaining the influence of the user cluster according to the analysis result; and recommending according to the ranking of the influence of the user cluster.
Optionally, the clustering algorithm is a k-means algorithm.
Optionally, extracting images of each frame of the video and extracting features may be implemented by one of a Scale-invariant feature transform (SIFT), a gray level co-occurrence matrix method, a fourier power spectrum method, and a deep learning model.
Optionally, analyzing the interaction relationship between the users in the user cluster in the video cluster specifically includes: forming a huge network G according to the interactive relation among users, V = { u1, u2, \8230;, un }, wherein n is the number of users, and calculating an influence transfer matrix W ∈ R according to the interactive information of the users on the video n×n ：
Wherein, C ij The number of times of commenting and/or praise on the video published by the user i for the user j, the number of times of continuing commenting or praise on the video after the user i commends or praise on the video for the user j, and k belongs to N (u) j ) All neighbor buddies for user j, i.e., the attention buddies. α, β are the importance weight values of different interactions, respectively, and W (i, j) describes the influence of user i on user j.
Optionally, obtaining the influence of the user cluster according to the analysis result specifically includes: i is t+1 ＝bWI t +(1-b)I 0 Wherein, I t ∈R 1×n The influence values of all users at the time t are obtained; t =0, I 0 Each element value of (1/n); b is an adjustable parameter.
Optionally, obtaining the influence of the user cluster according to the analysis result specifically includes: judgment of I t And I t+1 Whether the difference satisfies a convergence condition, and if so, outputting I t 。
Optionally, the convergence condition is a set threshold or a threshold range.
The embodiment of the invention provides a video recommendation device, which comprises an acquisition unit, a recommendation unit and a recommendation unit, wherein the acquisition unit is used for collecting a plurality of user behavior data in a preset time; the storage unit is used for storing the user behavior data sample; the first clustering unit is used for clustering the video content of the user behavior data sample by adopting a clustering algorithm to obtain at least one user cluster with similar interest; the statistical unit is used for counting videos watched by the user cluster users with similar interests; the image unit is used for extracting images of each frame of the video and extracting features; the second clustering unit is used for clustering the characteristics of each frame of image of the video by adopting a clustering algorithm to obtain at least one video cluster; the analysis unit is used for analyzing the interactive relation among the users in the user cluster in the video clustering and obtaining the influence of the user cluster according to the analysis result; and the recommending unit is used for recommending according to the ranking of the influence of the user cluster.
Optionally, the clustering algorithm adopted by the first clustering unit and the second clustering unit is a k-means algorithm.
Optionally, the image unit extracts each frame of image of the video and extracts features, and may be implemented by one of a Scale-invariant feature transform (SIFT) algorithm, a gray level co-occurrence matrix method, a fourier power spectrum method, and a deep learning model.
The embodiment of the invention provides a video recommendation method and device, which can find high-quality videos meeting the user interest for users by performing video recommendation according to user behaviors, video content clustering results and user influence results, and improve the user experience, user stickiness, user retention rate and liveness of video products.
Drawings
Fig. 1 is a flowchart illustrating a video recommendation method according to a first embodiment of the present invention;
fig. 2 is a schematic application diagram of a video recommendation method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of user interaction of a video recommendation method according to an embodiment of the present invention;
FIG. 4 is a block diagram of a video recommender according to a first embodiment of the present invention;
Detailed Description
The embodiment of the invention provides a video recommendation method and device, and the video recommendation method and device are shown in the figure 1, which is a schematic flow chart of the video recommendation method and device, and are used for finding out a part of people with high influence, preferentially recommending proper videos to the people, and recommending the videos to friends of the people after the people have some interaction on the videos. Under the condition of giving the total recommended amount, more comments or praise of the user on the video can be obtained through the releasing mode, the video can be enabled to have a deeper impression to the user, the user experience and interaction frequency are improved, and the user retention rate is improved. The method comprises the following steps:
s101, collecting a plurality of user behavior data in preset time, and establishing a user behavior data sample;
s103, storing the user behavior data sample;
in some embodiments, the original behavior data of the user is firstly obtained within a preset time, then the original behavior data is formatted and arranged according to a pre-established data specification to form a new user behavior data sample which accords with the specification, and finally a data storage label and a classification catalog are established for the complete user behavior data sample which accords with the specification, and the user behavior data sample data is stored.
S105, clustering the video content of the user behavior data sample by adopting a clustering algorithm to obtain at least one user cluster with similar interest;
in some embodiments, the clustering algorithm is a k-means algorithm. For example, on a short video platform, IThey may collect a record of the user's recent viewing of the video. These records reflect the user's preferences for the video. For example, on a video platform, assuming that we have four short videos and three users, user 1 watches videos 1, 2 and 3, user 2 watches short video 4, and user 3 watches videos 1 and 2, we can use table 1 to represent the historical behavior records of the users. Where the number 1 indicates that the user watched the video and 0 indicates that there was no activity. If we use a vector to quantify the user's recordings on all videos, users 1, 2, 3 can be represented as x, respectively 1 ＝[1110]、x 2 ＝[0001]And x 3 ＝[1100]。
Based on the vectorized user behavior records, the similarity between users can be calculated, and then the clustering algorithm is used for finding out users with similar interests. There are many choices for the similarity function, such as Cosine similarity between vectors, pearson correlation coefficient, etc. By means of a clustering algorithm, e.g. k-means, it can be deduced from the calculation method that user 1 and user 3 are users of the same type and user 2 is a user of another type.
The K-means algorithm is a hard clustering algorithm, is a typical target function clustering method based on a prototype, takes a certain distance from a data point to the prototype as an optimized target function, and obtains an adjustment rule of iterative operation by using a function extremum solving method. The K-means algorithm takes Euclidean distance as similarity measure, and the optimal classification of the central vector V corresponding to a certain initial cluster is solved, so that the evaluation index J is minimum. The algorithm uses a sum of squared errors criterion function as a clustering criterion function.
S107, counting videos watched by the user cluster users with similar interests;
s109, extracting each frame of image of the video and extracting features;
in some embodiments, extracting each frame of image of the video and extracting features may be performed by one of Scale-invariant feature transform (SIFT), gray-Scale co-occurrence matrix, fourier power spectroscopy, and deep learning models.
S111, clustering the characteristics of each frame of image of the video by adopting a clustering algorithm to obtain at least one video cluster;
s113, analyzing the interaction relation among the users in the user cluster in the video clustering, and obtaining the influence of the user cluster according to the analysis result;
in some embodiments, analyzing the interaction relationship among the users in the user cluster in the video cluster specifically includes: forming a huge network G according to the interactive relation among users, V = { u1, u2, \8230;, un }, wherein n is the number of users, and calculating an influence transfer matrix W ∈ R according to the interactive information of the users on the video n×n ：
Wherein, C ij The number of times of commenting and/or commenting on the video published by the user i for the user j, the number of times of continuing commenting or commenting on the video after the user i comments or commends on the video for the user j, and k belongs to N (u) j ) All neighbor buddies for user j, i.e., the attention buddies. α, β are the importance weight values of different interactions, respectively, and W (i, j) describes the influence of user i on user j.
The influence of the user cluster obtained according to the analysis result is specifically as follows: i is t+1 ＝bWI t +(1-b)I 0 In which I t ∈R 1×n The influence values of all users at the time t are obtained; t =0, I 0 Each element value of (1/n); b is an adjustable parameter.
The influence of the user cluster obtained according to the analysis result is specifically as follows: judgment of I t And I t+1 Whether the difference satisfies a convergence condition, and if so, outputting I t . The convergence condition is a set threshold or a threshold range.
Referring to fig. 2, it is assumed that the short video interactive network is composed of 4 users, and the interaction between them is shown in fig. 1, i.e., nodes U1, \8230, U4 represents 4 users, and directed edges represent the interaction between users. For example, the directional side U4- > U1 represents the behavior of the user U4 to the user U1, and the two numbers on the side respectively represent that the user U4 has 2 praise on the short video posted by the user U1 and has 1 follow-up comment on the short video commented by the user U1. Setting α =0.5 and β =0.5, and then calculating the influence transfer matrix according to equation (1):
then initializing I 0 =0.85 (0.25 ). W, I 0 And b is substituted into t round iterative operation to obtain the final influence value I of each user t = (1.29, 1.33,0.87, 1.13). It can be seen that the influence of the user U2 is the largest and the influence of U3 is the smallest.
And S115, recommending according to the ranking of the influence of the user cluster.
In some embodiments, referring to fig. 3, for example, first clustering is performed according to the user watching behaviors, to find user clusters with similar interests, for example, a sports user cluster and an entertainment user cluster, and then clustering is performed according to the content on videos watched by the sports user cluster users, for example, to find videos related to football. And finally, finding out the subset of the users with larger influence according to the interactive behaviors of the users on each type of videos. Finally, when recommendation is carried out, the user is preferentially pushed with videos which are well appreciated and appreciated by the high-influence user.
Referring to fig. 4, a schematic diagram of a video recommendation device module according to a first embodiment of the present invention is shown, where the video recommendation device 10 may be any terminal with a display function, such as a mobile phone, a PDA (Personal Digital Assistant) or a tablet computer, a portable communication device, a computer, a smart television, or a wearable device, such as a smart band, or a wearable device embedded in a clothing accessory, such as clothes, jewelry, or the like. The video recommendation device comprises:
an acquiring unit 101, which collects a plurality of user behavior data within a predetermined time and establishes a user behavior data sample;
a storage unit 103, configured to store the user behavior data sample;
in some embodiments, the original behavior data of the user is firstly acquired, then the original behavior data is formatted according to a pre-established data specification to form a new user behavior data sample which accords with the specification, and finally a data storage label and a classification catalogue are established for the complete user behavior data sample which accords with the specification, and the user behavior data sample data is stored.
In some embodiments, the storage unit 103 is a storage medium having any one type of a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), a RAM, and a ROM (EEPROM, etc.).
The first clustering unit 105 is configured to cluster the video content of the user behavior data sample by using a clustering algorithm to obtain at least one user cluster with similar interests;
in some embodiments, the clustering algorithm is a k-means algorithm.
A statistic unit 107, configured to count videos that users of the user cluster with similar interests have watched;
in some embodiments, the statistical unit 107 may calculate similarity between users based on vectorized user behavior records, and then find users with similar interests using a clustering algorithm. There are many choices for the function of similarity, such as Cosine similarity between vectors, pearson correlation coefficient, etc.
An image unit 109, configured to extract an image of each frame of the video and extract features;
in some embodiments, the image unit 109 extracts each frame of image of the video and extracts features by using one of Scale-invariant feature transform (SIFT), gray level co-occurrence matrix method, fourier power spectrum method, and deep learning model.
The second clustering unit 111 is used for clustering the characteristics of each frame of image of the video by adopting a clustering algorithm to obtain at least one video cluster;
an analysis unit 113, configured to analyze an interaction relationship between users in the user cluster in the video cluster, and obtain an influence of the user cluster according to the analysis result;
in some embodiments, the analyzing unit 113 analyzes the interaction relationship among the users in the user cluster in the video cluster specifically as follows: forming a huge network G according to the interactive relation among users, V = { u1, u2, \8230;, un }, wherein n is the number of users, and calculating an influence transfer matrix W ∈ R according to the interactive information of the users on the video n×n ：
Wherein, C ij The number of times of commenting and/or praise on the video published by the user i for the user j, the number of times of continuing commenting or praise on the video after the user i commends or praise on the video for the user j, and k belongs to N (u) j ) All neighbor buddies for user j, i.e., the attention buddies. α, β are the importance weight values of different interactions, respectively, and W (i, j) describes the influence of user i on user j.
The influence of the user cluster obtained according to the analysis result is specifically as follows: i is t+1 ＝bWI t +(1-b)I 0 In which I t ∈R 1×n The influence values of all users at the time t are obtained; t =0, I 0 Each element value of (a) is equal to 1/n; b is an adjustable parameter.
The influence of the user cluster obtained according to the analysis result is specifically as follows: judgment of I t And I t+1 Whether the difference satisfies a convergence condition, and if so, outputting I t . The convergence condition is a set threshold or a threshold range.
Referring to fig. 2, it is assumed that the short video interactive network is composed of 4 users, and the interaction between them is shown in fig. 1, i.e., nodes U1, \8230, U4 represents 4 users, and directed edges represent the interaction between users. For example, the directed edge U4- > U1 represents the behavior of the user U4 to the user U1, and the two numbers on the edge respectively represent that the user U4 has 2 praise on the short video posted by the user U1 and has 1 follow-up comment on the short video commented by the user U1. Setting α =0.5 and β =0.5, and then calculating the influence transfer matrix according to equation (1):
post-initialization I 0 = 0.25, setting b =0.85. We will turn W, I 0 And b is substituted into t round iterative operation to obtain the final influence value I of each user t = (1.29, 1.33,0.87, 1.13). It can be seen that the influence of the user U2 is the largest and the influence of U3 is the smallest.
And the recommending unit 115 is configured to recommend according to the ranking of the influence of the user cluster.
Referring to fig. 3, for example, first, clustering is performed according to the watching behaviors of users to find user clusters with similar interests, for example, an entertainment user cluster and an entertainment user cluster, and then, clustering is performed on videos watched by users in a sports user cluster according to contents, for example, the videos are classified as entertainment videos related to the country or the europe and the united states. And finally, finding out the subset of the users with larger influence according to the interactive behaviors of the users on each type of videos. Finally, when recommendation is carried out, the user is preferentially pushed with videos which are well appreciated and appreciated by the high-influence user.
If the above functions of this embodiment are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer readable storage medium, that is, the embodiments of the present invention may be embodied in the form of software products, which include several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present invention.
On the basis of the above, the above mentioned embodiments are only examples of the present invention, and not intended to limit the scope of the present invention, and all equivalent structures or equivalent flow transformations made by using the contents of the present specification and the drawings, such as the combination of technical features between the embodiments, or the direct or indirect application to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A video recommendation method, comprising the steps of:
collecting a plurality of user behavior data within a preset time, and establishing a user behavior data sample;
storing the user behavior data samples;
clustering the video content of the user behavior data sample by adopting a clustering algorithm to obtain at least one user cluster with similar interest;
counting videos watched by the user cluster users with similar interests;
extracting each frame of image of the video and extracting features;
clustering the characteristics of each frame of image of the video by adopting a clustering algorithm to obtain at least one video cluster;
analyzing the interactive relation among the users in the user cluster in the video cluster, and obtaining the influence of the user cluster according to the analysis result;
and recommending according to the ranking of the influence of the user cluster.
2. The video recommendation method of claim 1, wherein said clustering algorithm is a k-means algorithm.
3. The video recommendation method of claim 1, wherein extracting images of each frame of the video and extracting features are implemented by one of Scale-invariant feature transform (SIFT), gray level co-occurrence matrix, fourier power spectrum, and deep learning model.
4. The video recommendation method of claim 1, wherein analyzing the interaction relationship among the users in the user cluster in the video cluster specifically comprises: forming a huge network G according to the interactive relation among users, V = { u1, u2, \8230;, un }, n is the number of the users, and calculating an influence transfer matrix W epsilon R according to the interactive information of the users on the video n ×n ：
Wherein, C ij For the number of comments and/or praise of user j to the video published by user i, A ij For the number of times user j continues to comment or like after user i has commented or liked a certain video,for all the neighboring friends of the user j, namely the attention friends, α and β are respectively importance weighted values of different interactions, and W (i, j) describes the influence of the user i on the user j.
5. The video recommendation method of claim 4, wherein obtaining the influence of the user cluster according to the analysis result specifically comprises: i is t+1 ＝bWI t +(1-b)I 0 Wherein, I t ∈R 1×n The influence values of all users at the moment t are obtained; t =0, I 0 Each element value of (a) is equal to 1/n; b is an adjustable parameter.
6. The video recommendation method of claim 5, wherein obtaining the influence of the user cluster according to the analysis result specifically comprises: judgment of I t And I t+1 Whether the difference satisfies a convergence condition, and if so, outputting I t 。
7. The video recommendation method of claim 6, wherein the convergence condition is a set threshold or a threshold range.
8. A video recommendation apparatus, characterized in that the video recommendation apparatus comprises:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for collecting a plurality of user behavior data in a preset time and establishing user behavior data samples;
the storage unit is used for storing the user behavior data sample;
the first clustering unit is used for clustering the video content of the user behavior data sample by adopting a clustering algorithm to obtain at least one user cluster with similar interest;
the statistical unit is used for counting videos watched by the user cluster users with similar interests;
the image unit is used for extracting each frame of image of the video and extracting features;
the second clustering unit is used for clustering the characteristics of each frame of image of the video by adopting a clustering algorithm to obtain at least one video cluster;
the analysis unit is used for analyzing the interactive relation among the users in the user cluster in the video clustering and obtaining the influence of the user cluster according to the analysis result;
and the recommending unit is used for recommending according to the ranking of the influence of the user cluster.
9. The video recommendation device of claim 8, wherein the clustering algorithm employed by the first clustering unit and the second clustering unit is a k-means algorithm.
10. The video recommendation apparatus of claim 8, wherein the image unit extracts each frame of image of the video and extracts features, and is implemented by one of Scale-invariant feature transform (SIFT), gray level co-occurrence matrix method, fourier power spectrum method, and deep learning model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710997564.0A CN107894998B (en) | 2017-10-24 | 2017-10-24 | Video recommendation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710997564.0A CN107894998B (en) | 2017-10-24 | 2017-10-24 | Video recommendation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107894998A true CN107894998A (en) | 2018-04-10 |
| CN107894998B CN107894998B (en) | 2019-04-26 |
Family
ID=61803748
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710997564.0A Active CN107894998B (en) | 2017-10-24 | 2017-10-24 | Video recommendation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107894998B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109214374A (en) * | 2018-11-06 | 2019-01-15 | 北京达佳互联信息技术有限公司 | Video classification methods, device, server and computer readable storage medium |
| CN109218769A (en) * | 2018-09-30 | 2019-01-15 | 武汉斗鱼网络科技有限公司 | A kind of recommended method and relevant device of direct broadcasting room |
| CN109242030A (en) * | 2018-09-21 | 2019-01-18 | 京东方科技集团股份有限公司 | Draw single generation method and device, electronic equipment, computer readable storage medium |
| CN109740063A (en) * | 2019-01-17 | 2019-05-10 | 北京奇艺世纪科技有限公司 | Information recalls, information cluster method, device and equipment |
| CN110992137A (en) * | 2019-11-27 | 2020-04-10 | 上海风秩科技有限公司 | Product recommendation method and device, electronic equipment and storage medium |
| CN110996144A (en) * | 2019-12-19 | 2020-04-10 | 哈尔滨工业大学 | Television user viewing preference analysis method |
| TWI702844B (en) * | 2018-07-04 | 2020-08-21 | 大陸商騰訊科技（深圳）有限公司 | Method, device, apparatus, and storage medium of generating features of user |
| CN112287166A (en) * | 2020-09-23 | 2021-01-29 | 山东师范大学 | Movie recommendation method and system based on improved deep belief network |
| CN112449240A (en) * | 2020-11-10 | 2021-03-05 | 深圳市易平方网络科技有限公司 | User loss prediction method and terminal based on Internet television use behaviors |
| CN117708436A (en) * | 2024-02-05 | 2024-03-15 | 福州掌中云科技有限公司 | Network literature short-play recommendation management system based on big data |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102334118A (en) * | 2010-11-29 | 2012-01-25 | 华为技术有限公司 | Promoting method and system for personalized advertisement based on interested learning of user |
| US20140010518A1 (en) * | 2012-06-25 | 2014-01-09 | Sony Corporation | Video recording apparatus, information processing system, information processing method, and recording medium |
| CN103945238A (en) * | 2014-05-07 | 2014-07-23 | 北京邮电大学 | Community detection method based on user behaviors |
| CN104408210A (en) * | 2014-12-31 | 2015-03-11 | 合一网络技术(北京)有限公司 | Video recommendation method based on opinion leaders |
| CN103377242B (en) * | 2012-04-25 | 2016-06-22 | Tcl集团股份有限公司 | User behavior analysis method, analyzing and predicting method and television program push system |
| CN106446078A (en) * | 2016-09-08 | 2017-02-22 | 乐视控股（北京）有限公司 | Information recommendation method and recommendation apparatus |
| CN106952166A (en) * | 2016-01-07 | 2017-07-14 | 腾讯科技（深圳）有限公司 | The user force evaluation method and device of a kind of social platform |
| CN107038213A (en) * | 2017-02-28 | 2017-08-11 | 华为技术有限公司 | A kind of method and device of video recommendations |
-
2017
- 2017-10-24 CN CN201710997564.0A patent/CN107894998B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102334118A (en) * | 2010-11-29 | 2012-01-25 | 华为技术有限公司 | Promoting method and system for personalized advertisement based on interested learning of user |
| CN103377242B (en) * | 2012-04-25 | 2016-06-22 | Tcl集团股份有限公司 | User behavior analysis method, analyzing and predicting method and television program push system |
| US20140010518A1 (en) * | 2012-06-25 | 2014-01-09 | Sony Corporation | Video recording apparatus, information processing system, information processing method, and recording medium |
| CN103945238A (en) * | 2014-05-07 | 2014-07-23 | 北京邮电大学 | Community detection method based on user behaviors |
| CN104408210A (en) * | 2014-12-31 | 2015-03-11 | 合一网络技术(北京)有限公司 | Video recommendation method based on opinion leaders |
| CN106952166A (en) * | 2016-01-07 | 2017-07-14 | 腾讯科技（深圳）有限公司 | The user force evaluation method and device of a kind of social platform |
| CN106446078A (en) * | 2016-09-08 | 2017-02-22 | 乐视控股（北京）有限公司 | Information recommendation method and recommendation apparatus |
| CN107038213A (en) * | 2017-02-28 | 2017-08-11 | 华为技术有限公司 | A kind of method and device of video recommendations |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI702844B (en) * | 2018-07-04 | 2020-08-21 | 大陸商騰訊科技（深圳）有限公司 | Method, device, apparatus, and storage medium of generating features of user |
| CN109242030A (en) * | 2018-09-21 | 2019-01-18 | 京东方科技集团股份有限公司 | Draw single generation method and device, electronic equipment, computer readable storage medium |
| CN109218769A (en) * | 2018-09-30 | 2019-01-15 | 武汉斗鱼网络科技有限公司 | A kind of recommended method and relevant device of direct broadcasting room |
| CN109218769B (en) * | 2018-09-30 | 2021-01-01 | 武汉斗鱼网络科技有限公司 | Recommendation method for live broadcast room and related equipment |
| CN109214374A (en) * | 2018-11-06 | 2019-01-15 | 北京达佳互联信息技术有限公司 | Video classification methods, device, server and computer readable storage medium |
| CN109740063A (en) * | 2019-01-17 | 2019-05-10 | 北京奇艺世纪科技有限公司 | Information recalls, information cluster method, device and equipment |
| CN110992137A (en) * | 2019-11-27 | 2020-04-10 | 上海风秩科技有限公司 | Product recommendation method and device, electronic equipment and storage medium |
| CN110992137B (en) * | 2019-11-27 | 2023-09-01 | 上海风秩科技有限公司 | Product recommendation method and device, electronic equipment and storage medium |
| CN110996144A (en) * | 2019-12-19 | 2020-04-10 | 哈尔滨工业大学 | Television user viewing preference analysis method |
| CN112287166A (en) * | 2020-09-23 | 2021-01-29 | 山东师范大学 | Movie recommendation method and system based on improved deep belief network |
| CN112287166B (en) * | 2020-09-23 | 2023-03-07 | 山东师范大学 | Movie recommendation method and system based on improved deep belief network |
| CN112449240A (en) * | 2020-11-10 | 2021-03-05 | 深圳市易平方网络科技有限公司 | User loss prediction method and terminal based on Internet television use behaviors |
| CN112449240B (en) * | 2020-11-10 | 2022-12-06 | 深圳市易平方网络科技有限公司 | User loss prediction method and terminal based on Internet television use behaviors |
| CN117708436A (en) * | 2024-02-05 | 2024-03-15 | 福州掌中云科技有限公司 | Network literature short-play recommendation management system based on big data |
| CN117708436B (en) * | 2024-02-05 | 2024-04-26 | 福州掌中云科技有限公司 | Network literature short-play recommendation management system based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107894998B (en) | 2019-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107894998A (en) | Video recommendation method and device | |
| CN110321422B (en) | Method for training model on line, pushing method, device and equipment | |
| Segalin et al. | What your Facebook profile picture reveals about your personality | |
| CN108875022B (en) | Video recommendation method and device | |
| CN106326391B (en) | Multimedia resource recommendation method and device | |
| WO2017181612A1 (en) | Personalized video recommendation method and device | |
| WO2020143156A1 (en) | Hotspot video annotation processing method and apparatus, computer device and storage medium | |
| CN109033149B (en) | Information recommendation method and device, server and storage medium | |
| US20200349385A1 (en) | Multimedia resource matching method and apparatus, storage medium, and electronic apparatus | |
| US10460174B2 (en) | System and methods for analysis of user-associated images to generate non-user generated labels and utilization of the generated labels | |
| CN103365936A (en) | Video recommendation system and method thereof | |
| CN107305557A (en) | Content recommendation method and device | |
| CN104182449A (en) | System and method for personalized video recommendation based on user interests modeling | |
| CN110008397B (en) | Recommendation model training method and device | |
| CN108959323B (en) | Video classification method and device | |
| CN106446266A (en) | Method for recommending favorite content to user and content recommending system | |
| Hao et al. | Emotion-aware video QoE assessment via transfer learning | |
| CN101668176A (en) | Multimedia content-on-demand and sharing method based on social interaction graph | |
| CN107247919A (en) | The acquisition methods and system of a kind of video feeling content | |
| KR20170079429A (en) | A clustering based collaborative filtering method with a consideration of users' features and movie recommendation system using thereof | |
| CN111597446B (en) | Content pushing method and device based on artificial intelligence, server and storage medium | |
| Narassiguin et al. | Data Science for Influencer Marketing: feature processing and quantitative analysis | |
| CN112685596B (en) | Video recommendation method and device, terminal and storage medium | |
| CN108804492B (en) | Method and device for recommending multimedia objects | |
| Wasilewski et al. | Are you reaching your audience? Exploring item exposure over consumer segments in recommender systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |