CN108347652B - Method and system for recommending IPTV live broadcast channel by using artificial neural network - Google Patents

Method and system for recommending IPTV live broadcast channel by using artificial neural network Download PDF

Info

Publication number
CN108347652B
CN108347652B CN201810157400.1A CN201810157400A CN108347652B CN 108347652 B CN108347652 B CN 108347652B CN 201810157400 A CN201810157400 A CN 201810157400A CN 108347652 B CN108347652 B CN 108347652B
Authority
CN
China
Prior art keywords
user
channel
watching
data
sequence
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.)
Active
Application number
CN201810157400.1A
Other languages
Chinese (zh)
Other versions
CN108347652A (en
Inventor
杨灿
任思璇
刘勇
韩国强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China University of Technology SCUT
Original Assignee
South China University of Technology SCUT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by South China University of Technology SCUT filed Critical South China University of Technology SCUT
Priority to CN201810157400.1A priority Critical patent/CN108347652B/en
Publication of CN108347652A publication Critical patent/CN108347652A/en
Application granted granted Critical
Publication of CN108347652B publication Critical patent/CN108347652B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
    • H04N21/44213—Monitoring of end-user related data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Biophysics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Social Psychology (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a method and a system for recommending IPTV live channels by using an artificial neural network, wherein the method comprises the following steps: s1, selecting a sliding window of the training data, and screening out the original training data of each user in the time period; s2, performing data cleaning on the extracted original training data; s3, training the cleaned data, obtaining training models of respective devices according to each device number, predicting by using the corresponding training models of each device in t +1 days, acquiring information of a current channel watched by a user in real time by the devices, sending the information into the trained models for prediction, and pushing the information to the corresponding devices; s4, when the current date is over, the end date of the original training data window is set as t +1, and the step S1 is repeated. The method uses three methods to train the model for each user, achieves better effect, improves the user experience of IPTV, and is easy to popularize and use.

Description

Method and system for recommending IPTV live broadcast channel by using artificial neural network
Technical Field
The invention relates to the technical field of IPTV and the technical field of recommendation, in particular to a method and a system for recommending IPTV live channels by using an artificial neural network.
Background
With the continuous progress of the technology, the television industry is developed vigorously, and television program contents are enriched day by day and develop towards diversification. Among them, IPTV is highly popular among a large number of users due to its clear interactive characteristics. Meanwhile, a problem arises in that, after enjoying the enjoyment of watching a program, a user needs to constantly switch among channels as much as the sea to select a next channel of interest, long-time operation reduces user experience, and resources are wasted. Although some systems have designed a search function, the user still has no definite concept for the program he wants to watch, and is limited to search a few programs provided by the television station, and the search function is not fully utilized. Therefore, channel recommendation becomes an important requirement for users and a problem to be solved urgently.
The chinese patent application No. 201610258946.7 discloses a video recommendation system based on a television set-top box, in which a recommendation engine uses collaborative filtering and content filtering in a mixed manner, analyzes the interest of a user from the historical behavior of the user, and recommends a video meeting the interest of the user to the user. In the video recommendation system and method disclosed in the chinese patent application No. 201310684807.7, the system needs to obtain multi-source data such as user personal information and social network information through an information obtaining module, so as to analyze the personality, emotion and taste of the user, thereby recommending channels of interest to the user. The television channel recommendation system and the recommendation method disclosed in chinese patent application No. 200410083188.7 calculate the recommendation score of each channel according to the recorded watching time and the channel number, and recommend the channel with the maximum score to the user. The chinese patent application No. 201110203368.4 discloses a television channel recommendation system and method, which includes a front end system and a terminal system, wherein the front end system includes a user behavior acquisition server and an intelligent analysis server, the intelligent analysis server performs a series of processing on the information to obtain a user behavior trajectory, further processes the user behavior trajectory to obtain summary information, and then recommends programs matched with the summary information in a program library to the user. However, most of the prior art recommends IPTV on-demand channels, and needs too many user attributes, such as program information watched by the user, which has a narrow application range.
Disclosure of Invention
The invention aims to provide a method for recommending IPTV live channels by using an artificial neural network, aiming at the defects of the prior art, the method integrates various artificial neural networks to recommend the IPTV live channels to users, does not need too many user attributes, and has wider application range.
Another object of the present invention is to provide a system for recommending IPTV live channels by using an artificial neural network.
The purpose of the invention can be realized by the following technical scheme:
a method for recommending IPTV live broadcast channels by using an artificial neural network comprises the following steps:
s1, selecting sliding window days delta T, and taking data in a [ T-delta T, T ] time window as original training data, wherein T represents the ending date of the training data, the time T is not more than or equal to the current watching date of a user, the original training data comprises the following data structure of < equipment number, the watching time, the watching channel and the watching duration >, and the equipment number is not limited to the equipment number of the set top box;
s2, cleaning the extracted original training data to remove noise data generated by the user due to fast channel switching, wherein the cleaned data can represent the behavior characteristics of the user;
s3, training the cleaned data, obtaining training models of respective devices according to each device number, predicting by using the corresponding training models of each device in t +1 days, acquiring information of a current channel watched by a user in real time by the devices, sending the information into the trained models for prediction, and pushing the information to the corresponding devices;
s4, when the current date is over, the end date of the original training data window is set as t +1, and the step S1 is repeated.
Further, the specific process of step S2 is: deleting the records of the original training data with the user watching time length less than 10 seconds and more than 3 hours, removing irrelevant attributes of the original training data when the user watches, wherein the reserved watching attributes are < equipment number, current watching channel, next watching channel, date >, channel watched by the equipment number in the date and next channel watched by the equipment number, the date is obtained from the watching time in the original training data, and the cleaned data are arranged according to the ascending order of the watching time of the user.
Further, training the cleaned data in step S3 can take three methods: a recurrent neural network recommendation method, an inverse neural network recommendation method or a multi-network cold and hot channel hybrid recommendation method.
Further, in step S3, a recurrent neural network recommendation method is used for training the cleaned data, and the specific process is as follows:
s3.1, will clearDividing the data after washing according to the number of each device, and obtaining the watching sequence alpha of each user ═ C1，C2，…，CnThe watching sequence is arranged according to the watching time of the user, CiIndicating the channel that the device watched the ith time within the training window Δ T;
s3.2, reconstructing the watching sequence of each user, namely, the watching sequence alpha is reconstructed from C1To CmSliding backwards according to a sequence length of m, where C1To CmForm a first sequence, Cm+1For the label of the sequence, the watching sequence alpha is finally divided into n-m +1 sequences with the length of m, the label of each sequence is the next channel of the last channel of the sequence, and the divided sequences are arranged according to the watching sequence of the user;
s3.3, inputting each sequence into a recurrent neural network for training according to a user, and finally obtaining a prediction model of each sequence;
s3.4, the equipment collects the watching information of the user in real time in t +1 days, and the equipment comprises the following structures: and inputting the current channel into the trained equipment prediction model with the sequence length of m, and pushing the channel number output by the prediction model to corresponding equipment.
Further, in step S3, the training of the cleaned data employs an inverse neural network recommendation method, which specifically includes:
s3.1, dividing the cleaned data according to the number of each device, and obtaining a watching sequence alpha of each user ═ C1，C2，…，CnThe watching sequence is arranged according to the watching time of the user, CiIndicating the channel that the device watched the ith time within the training window Δ T;
s3.2, constructing a training data set, and labeling a label for a watching sequence of each user, wherein the channel CiIs labeled with Ci+1，Ci+1Watching channel C for user UiFor each user, the viewing sequence α ═ C for each user1，C2，…，CnThe sequence of labels L ═ C is obtained2，C3…，Cn+1I.e. viewing channel C in sequence a1The corresponding label is C2Channel CnThe corresponding label is Cn+1If C isnFor the last channel viewed that day, label Cn+1Set to 0, indicating shutdown;
s3.3, inputting the watching sequence alpha of each user and the corresponding label sequence L into a reverse neural network to obtain a training model for each user;
s3.4, the equipment collects the watching information of the user in real time in t +1 days, and the equipment comprises the following structures: and inputting the current channel into the trained equipment prediction model according to the equipment number, and pushing the channel number output by the model to corresponding equipment.
Further, in step S3, the training of the cleaned data employs a multi-network cold-hot channel hybrid recommendation method, which includes the following specific steps:
s3.1, dividing the cleaned data according to the equipment number, and counting the time window [ T-delta T, T ] of each user]Setting the threshold value of the Cold channel as rho% of the viewing frequency of each channel, and identifying the channel with the viewing frequency less than or equal to rho% as the Cold channel to obtain a Cold channel set Coldi＝{C1，C2……,Cx}，ColdiA cold channel set representing the user i, and a channel with the watching frequency greater than rho percent is determined as a Hot channel, so that a Hot channel set Hot is obtainedi＝{C1，C2……,Cy}，HotiA hot channel set representing user i;
s3.2, training a cold channel prediction model and a hot channel prediction model of each user, namely obtaining the cold channel prediction model of each user by adopting an inverse neural network method aiming at the cold channel of each user; aiming at the hot channel of each user, a recurrent neural network method is adopted to obtain a hot channel prediction model of each user;
s3.3, the equipment collects the user watching information in real time in t +1 days, and the equipment comprises the following structures:<the number of the device is set to be,current channel, date>Inputting the current information into the trained equipment model, and if the currently watched channel is in the Cold channel set Cold of the useriIf so, predicting by using a cold channel prediction model of the user; if the currently watched channel is in the Hot channel set Hot of the useriIf so, predicting by using a popular channel prediction model of the user; and pushing the predicted first N channels to the corresponding users.
Further, the method adopts a buffer to store the recommended channel for each user every time, and if the user is a starting user, the channel which is stored in the buffer and is recommended last time is recommended for the user.
The other purpose of the invention can be realized by the following technical scheme:
a system for recommending IPTV live channels by using an artificial neural network comprises a data cleaning module, a recommending module and a pushing module, wherein original training data are recommended through the recommending module after being cleaned in the data cleaning module, the recommending module comprises a cold and hot channel discriminator for discriminating cold and hot channels of each user, a trainer for training the cleaned original training data and a predictor for predicting the hot channels of each user in real time, and a recommending result is sent to the recommending module to be pushed for the user.
Compared with the prior art, the invention has the following advantages and beneficial effects:
according to the method for recommending the IPTV live broadcast channel by using the artificial neural network, disclosed by the invention, the training model can be established through the long-term and short-term memory network only by acquiring the channel watching record of the user without other video data and description, so that a recommended channel list which is possibly interested in the user is provided in real time, and the user experience is greatly improved; in addition, the personalized real-time channel recommendation system provided by the invention continuously updates the user viewing data by using the sliding window, so that a new recommendation model is established, the channels which the user is interested in can be recommended to the user in real time, and the hit rate and the user experience are improved.
Drawings
Fig. 1 is a flowchart of a method for recommending an IPTV live broadcast channel by using an artificial neural network according to the present invention.
FIG. 2 is a schematic diagram of the present invention training the cleaned data using recurrent neural network recommendation.
FIG. 3 is a schematic diagram of the present invention training the cleaned data using the inverse neural network recommendation method.
FIG. 4 is a schematic diagram of the present invention for training data after cleaning by using a multi-network hot and cold channel mix recommendation method.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited thereto.
Example 1:
the embodiment provides a method for recommending an IPTV live channel by using an artificial neural network, a flowchart of the method is shown in fig. 1, and the method includes the following steps:
s1, selecting sliding window days delta T, and taking data in a [ T-delta T, T ] time window as original training data, wherein T represents the ending date of the training data, the time T is not more than or equal to the current watching date of a user, the original training data comprises the following data structure of < equipment number, the watching time, the watching channel and the watching duration >, and the equipment number is not limited to the equipment number of the set top box;
s2, cleaning the extracted original training data to remove noise data generated by the user due to fast channel switching, wherein the cleaned data can represent the behavior characteristics of the user;
the specific process is as follows: deleting the records of the original training data with the user watching time length less than 10 seconds and more than 3 hours, removing irrelevant attributes of the original training data when the user watches, wherein the reserved watching attributes are < equipment number, current watching channel, next watching channel, date >, channel watched by the equipment number in the date and next channel watched by the equipment number, the date is obtained from the watching time in the original training data, and the cleaned data are arranged according to the ascending order of the watching time of the user.
S3, training the cleaned data, obtaining training models of respective devices according to each device number, predicting by using the corresponding training models of each device in t +1 days, acquiring information of a current channel watched by a user in real time by the devices, sending the information into the trained models for prediction, and pushing the information to the corresponding devices;
wherein, training the cleaned data can adopt three methods: a recurrent neural network recommendation method, an inverse neural network recommendation method or a multi-network cold and hot channel hybrid recommendation method.
A schematic diagram of the recurrent neural network recommendation method is shown in fig. 2, and the specific process is as follows:
s3.1, dividing the cleaned data according to the number of each device, and obtaining a watching sequence alpha of each user ═ C1，C2，…，CnThe watching sequence is arranged according to the watching time of the user, CiIndicating the channel that the device watched the ith time within the training window Δ T;
s3.2, reconstructing the watching sequence of each user, namely, the watching sequence alpha is reconstructed from C1To CmSliding backwards according to a sequence length of m, where C1To CmForm a first sequence, Cm+1For the label of the sequence, the watching sequence alpha is finally divided into n-m +1 sequences with the length of m, the label of each sequence is the next channel of the last channel of the sequence, and the divided sequences are arranged according to the watching sequence of the user;
s3.3, inputting each sequence into a recurrent neural network for training according to a user, and finally obtaining a prediction model of each sequence;
s3.4, the equipment collects the watching information of the user in real time in t +1 days, and the equipment comprises the following structures: and inputting the current channel into the trained equipment prediction model with the sequence length of m, and pushing the channel number output by the prediction model to corresponding equipment.
A schematic diagram of the method for recommending the inverse neural network is shown in fig. 3, and the specific process is as follows:
s3.1, dividing the cleaned data according to the number of each device, and obtaining a watching sequence alpha of each user ═ C1，C2，…，CnThe watching sequence is arranged according to the watching time of the user, CiIndicating the channel that the device watched the ith time within the training window Δ T;
s3.2, constructing a training data set, and labeling a label for a watching sequence of each user, wherein the channel CiIs labeled with Ci+1，Ci+1Watching channel C for user UiFor each user, the viewing sequence α ═ C for each user1，C2，…，CnThe sequence of labels L ═ C is obtained2，C3…，Cn+1I.e. viewing channel C in sequence a1The corresponding label is C2Channel CnThe corresponding label is Cn+1If C isnFor the last channel viewed that day, label Cn+1Set to 0, indicating shutdown;
s3.3, inputting the watching sequence alpha of each user and the corresponding label sequence L into a reverse neural network to obtain a training model for each user;
s3.4, the equipment collects the watching information of the user in real time in t +1 days, and the equipment comprises the following structures: and inputting the current channel into the trained equipment prediction model according to the equipment number, and pushing the channel number output by the model to corresponding equipment.
Fig. 4 shows a schematic diagram of a multi-network hot and cold channel mixed recommendation method, which specifically includes the following steps:
s3.1, dividing the cleaned data according to the equipment number, and counting the time window [ T-delta T, T ] of each user]Setting the threshold value of the Cold channel as rho% of the viewing frequency of each channel, and identifying the channel with the viewing frequency less than or equal to rho% as the Cold channel to obtain a Cold channel set Coldi＝{C1，C2……,Cx}，ColdiRepresenting a cold channel set of user i, channels with a viewing frequency greater than ρ% are identified as hot channelsWay, get Hot channel set Hoti＝{C1，C2……,Cy}，HotiA hot channel set representing user i;
s3.2, training a cold channel prediction model and a hot channel prediction model of each user, namely obtaining the cold channel prediction model of each user by adopting an inverse neural network method aiming at the cold channel of each user; aiming at the hot channel of each user, a recurrent neural network method is adopted to obtain a hot channel prediction model of each user;
s3.3, the equipment collects the user watching information in real time in t +1 days, and the equipment comprises the following structures:<device number, current channel, date>Inputting the current information into the trained equipment model, and if the currently watched channel is in the Cold channel set Cold of the useriIf so, predicting by using a cold channel prediction model of the user; if the currently watched channel is in the Hot channel set Hot of the useriIf so, predicting by using a popular channel prediction model of the user; and pushing the predicted first N channels to the corresponding users.
S4, when the current date is over, the end date of the original training data window is set as t +1, and the step S1 is repeated.
Example 2:
the embodiment provides a system for implementing the method for recommending the IPTV live channels by using the artificial neural network, which comprises a data cleaning module (101), a recommending module (102) and a pushing module (103), wherein original training data are recommended through the recommending module (102) after being cleaned in the data cleaning module (101), the recommending module (102) comprises a cold and hot channel discriminator (201) for discriminating cold and hot channels of each user, a trainer (202) for training the cleaned original training data and a predictor (203) for predicting the hot channels of each user in real time, and a recommending result is sent to the recommending module (102) for pushing the user.
The above description is only for the preferred embodiments of the present invention, but the protection scope of the present invention is not limited thereto, and any person skilled in the art can substitute or change the technical solution of the present invention and the inventive concept within the scope of the present invention, which is disclosed by the present invention, and the equivalent or change thereof belongs to the protection scope of the present invention.

Claims (4)

1. A method for recommending IPTV live broadcast channels by using an artificial neural network is characterized by comprising the following steps:
s1, selecting sliding window days delta T, and taking data in a [ T-delta T, T ] time window as original training data, wherein T represents the ending date of the training data, the time T is not more than or equal to the current watching date of a user, the original training data comprises the following data structure of < equipment number, the watching time, the watching channel and the watching duration >, and the equipment number is not limited to the equipment number of the set top box;
s2, cleaning the extracted original training data to remove noise data generated by the user due to fast channel switching, wherein the cleaned data can represent the behavior characteristics of the user;
the specific process is as follows: deleting the records of the original training data with the user watching time length less than 10 seconds and more than 3 hours, removing irrelevant attributes of the original training data when the user watches, wherein the reserved watching attributes are < equipment number, current watching channel, next watching channel, date >, channel watched by the equipment number in the date and next channel watched by the equipment number, the date is obtained from the watching time in the original training data, and the cleaned data are arranged according to the ascending order of the watching time of the user;
s3, training the cleaned data, obtaining training models of respective devices according to each device number, predicting by using the corresponding training models of each device in t +1 days, acquiring information of a current channel watched by a user in real time by the devices, sending the information into the trained models for prediction, and pushing the information to the corresponding devices;
in step S3, a recurrent neural network recommendation method is used for training the cleaned data, and the specific process is as follows:
s3.1, the data after washing are based onEach device number is divided, and a viewing sequence α ═ C is obtained for each user1，C2，…，CnThe watching sequence is arranged according to the watching time of the user, CiIndicating the channel that the device watched the ith time within the training window Δ T;
s3.2, reconstructing the watching sequence of each user, namely, the watching sequence alpha is reconstructed from C1To CmSliding backwards according to a sequence length of m, where C1To CmForm a first sequence, Cm+1For the label of the sequence, the watching sequence alpha is finally divided into n-m +1 sequences with the length of m, the label of each sequence is the next channel of the last channel of the sequence, and the divided sequences are arranged according to the watching sequence of the user;
s3.3, inputting each sequence into a recurrent neural network for training according to a user, and finally obtaining a prediction model of each sequence;
s3.4, the equipment collects the watching information of the user in real time in t +1 days, and the equipment comprises the following structures: the method comprises the steps of (1) inputting a current channel into a trained equipment prediction model by taking the sequence length as m, and pushing a channel number output by the prediction model to corresponding equipment, < equipment number, current channel, date >;
s4, when the current date is over, the end date of the original training data window is set as t +1, and the step S1 is repeated.
2. The method as claimed in claim 1, wherein the step S2 is as follows: deleting the records of the original training data with the user watching time length less than 10 seconds and more than 3 hours, removing irrelevant attributes of the original training data when the user watches, wherein the reserved watching attributes are < equipment number, current watching channel, next watching channel, date >, channel watched by the equipment number in the date and next channel watched by the equipment number, the date is obtained from the watching time in the original training data, and the cleaned data are arranged according to the ascending order of the watching time of the user.
3. The method of claim 1, wherein the method for recommending IPTV live channels by using artificial neural network comprises: the method includes that a buffer is used for storing recommended channels for each user every time, and if the user is a starting user, the channel which is stored in the buffer and is recommended last time is recommended for the user.
4. A system for implementing the method for recommending IPTV live channels by using the artificial neural network as claimed in claim 1, comprising a data cleaning module, a recommending module and a pushing module, wherein the original training data are recommended by the recommending module after being cleaned in the data cleaning module, the recommending module comprises a cold and hot channel discriminator for discriminating cold and hot channels of each user, a trainer for training the cleaned original training data and a predictor for predicting hot channels of each user in real time, and the recommending result is sent to the recommending module for pushing the user.
CN201810157400.1A 2018-02-24 2018-02-24 Method and system for recommending IPTV live broadcast channel by using artificial neural network Active CN108347652B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810157400.1A CN108347652B (en) 2018-02-24 2018-02-24 Method and system for recommending IPTV live broadcast channel by using artificial neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810157400.1A CN108347652B (en) 2018-02-24 2018-02-24 Method and system for recommending IPTV live broadcast channel by using artificial neural network

Publications (2)

Publication Number Publication Date
CN108347652A CN108347652A (en) 2018-07-31
CN108347652B true CN108347652B (en) 2020-01-14

Family

ID=62959631

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810157400.1A Active CN108347652B (en) 2018-02-24 2018-02-24 Method and system for recommending IPTV live broadcast channel by using artificial neural network

Country Status (1)

Country Link
CN (1) CN108347652B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110225361B (en) * 2019-06-03 2021-10-15 武汉瓯越网视有限公司 Live broadcast room recommendation method, storage medium, electronic device and system
CN113343097B (en) * 2021-06-24 2023-01-13 中山大学 Sequence recommendation method and system based on fragment and self-attention mechanism
CN113986954B (en) * 2021-12-30 2022-04-08 深圳市明源云科技有限公司 User event acquisition method and device, intelligent terminal and readable storage medium
CN114528434B (en) * 2022-01-19 2023-07-21 华南理工大学 IPTV live channel fusion recommendation method based on self-attention mechanism
CN116320511A (en) * 2023-02-03 2023-06-23 华南理工大学 Cross-domain fusion recommendation method based on graph convolution network

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102510529A (en) * 2011-09-22 2012-06-20 中国科学技术大学 Method for performing on-demand play quantity prediction and memory scheduling on programs
CN103491441A (en) * 2013-09-09 2014-01-01 东软集团股份有限公司 Recommendation method and system of live television programs
CN104539981A (en) * 2014-11-26 2015-04-22 四川长虹电器股份有限公司 Real-time hot television channel recommending system and method
CN105138541A (en) * 2015-07-08 2015-12-09 腾讯科技（深圳）有限公司 Audio fingerprint matching query method and device
CN105791913A (en) * 2016-04-27 2016-07-20 青岛海信传媒网络技术有限公司 Multi-user telephone configuration information setting method, multi-user telephone configuration information setting system and cloud server
CN106101839A (en) * 2016-06-20 2016-11-09 徐汕 A kind of method identifying that television user gathers

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020174429A1 (en) * 2001-03-29 2002-11-21 Srinivas Gutta Methods and apparatus for generating recommendation scores
US9535897B2 (en) * 2013-12-20 2017-01-03 Google Inc. Content recommendation system using a neural network language model
US9843837B2 (en) * 2015-08-03 2017-12-12 At&T Intellectual Property I, L.P. Cross-platform analysis

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102510529A (en) * 2011-09-22 2012-06-20 中国科学技术大学 Method for performing on-demand play quantity prediction and memory scheduling on programs
CN103491441A (en) * 2013-09-09 2014-01-01 东软集团股份有限公司 Recommendation method and system of live television programs
CN104539981A (en) * 2014-11-26 2015-04-22 四川长虹电器股份有限公司 Real-time hot television channel recommending system and method
CN105138541A (en) * 2015-07-08 2015-12-09 腾讯科技（深圳）有限公司 Audio fingerprint matching query method and device
CN105791913A (en) * 2016-04-27 2016-07-20 青岛海信传媒网络技术有限公司 Multi-user telephone configuration information setting method, multi-user telephone configuration information setting system and cloud server
CN106101839A (en) * 2016-06-20 2016-11-09 徐汕 A kind of method identifying that television user gathers

Also Published As

Publication number Publication date
CN108347652A (en) 2018-07-31

Similar Documents

Publication Publication Date Title
CN108347652B (en) Method and system for recommending IPTV live broadcast channel by using artificial neural network
CN109922373B (en) Video processing method, device and storage medium
US9888279B2 (en) Content based video content segmentation
CN103442262B (en) User watched behavior analysis method and system based on television terminal video program
WO2016054916A1 (en) Video content recommending and evaluating methods and devices
CN103747343B (en) The method and apparatus that resource is recommended at times
CN102244811A (en) System and method for recommending television channels
CN107247733B (en) Video clip watching popularity analysis method and system
CN102495873B (en) Video recommending method based on video affective characteristics and conversation models
CN108419117B (en) Device and method for playing video small window
CN105187941B (en) A kind of television terminal and control method of intelligent management favorites
WO2020007470A1 (en) Dynamic viewer prediction system for advertisement scheduling
US20160330526A1 (en) Method and System for Automatically Selecting Parts of a Video and/or Audio Media Content Based on Information Obtained from Social Networks
CN109361955A (en) Speed video broadcasting method and system based on point of interest
US20190342614A1 (en) Integral Program Content Distribution
CN103581752A (en) Intelligent set top box program recommendation method based on cloud platform and implement system thereof
CN102695086B (en) The content delivery method of interactive network protocol TV and device
CN103442264A (en) Viewing analysis method and system based on video monitoring
CN112784069B (en) IPTV content intelligent recommendation system and method
CN108153882A (en) A kind of data processing method and device
CN103475911A (en) Television information providing method and system based on video characteristics
CN110309753A (en) A kind of race process method of discrimination, device and computer equipment
CN111263183A (en) Singing state identification method and singing state identification device
CN108419088B (en) Channel recommendation method facing high-burstiness user request
CN103475912A (en) Viewing analysis method and system based on video characteristics and user information

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