CN106604137A - Method and apparatus for predicting video viewing time length - Google Patents
Method and apparatus for predicting video viewing time length Download PDFInfo
- Publication number
- CN106604137A CN106604137A CN201611249506.1A CN201611249506A CN106604137A CN 106604137 A CN106604137 A CN 106604137A CN 201611249506 A CN201611249506 A CN 201611249506A CN 106604137 A CN106604137 A CN 106604137A
- Authority
- CN
- China
- Prior art keywords
- video
- duration
- viewing
- user
- information
- 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
- H04N21/43—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
- H04N21/44204—Monitoring of content usage, e.g. the number of times a movie has been viewed, copied or the amount which has been watched
-
- 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
- H04N21/44222—Analytics of user selections, e.g. selection of programs or purchase activity
-
- 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
- H04N21/4508—Management of client data or end-user data
- H04N21/4532—Management of client data or end-user data involving end-user characteristics, e.g. viewer profile, preferences
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Social Psychology (AREA)
- Computer Networks & Wireless Communication (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The invention discloses a method and apparatus for predicting . The method includes the following steps: if a preset operation for triggering a prediction video viewing time length is detected, acquiring the identification information of a current video viewing user and the attribute information of the video which is currently played; the attribute information including identification information and type information; based on the identification information of the current video viewing user and the attribute information of the video, acquiring an initial viewing time length prediction value; based on the preset corresponding relationship between the type information of the video and the adjustment ratio between the type information of the video and the errors, acquiring the error adjustment ratio corresponding to the video; based on the initial viewing time length prediction value and the error adjustment ratio, computing a final viewing time length prediction value. The aforementioned method can accurately predict the viewing time length of a user for a certain video so as to provide basis for optimizing resource allocation for each service provider.
Description
Technical field
The invention belongs to video playback field, more particularly to a kind of method and device of prediction video-see duration.
Background technology
Network video-on-demand is a kind of important Internet service.User can be by PC, mobile phone and intelligence
The equipment such as TV, watch its video liked by the Internet trick play.Each ginseng in network video-on-demand ecological device
It is devoted to always for many years improving service performance with person (such as advertiser, videoconference client developer etc.), so as to improve user
Experience, increase advertising income expand more value-added services.
Prediction user is an important prerequisite for improving service performance to the viewing duration of current video, is that each service is carried
For the important evidence of business's optimized allocation of resources.The meaning of the prediction includes：Based on the prediction come optimization of video download algorithm；It is based on
The prediction is determining advertisement putting content and form；And based on the prediction estimating the user activity of single unit system.
However, prior art can only predict the program request amount of video, and whether prediction user clicks on certain video, and can not
Viewing duration of the user to certain video is predicted directly.
The content of the invention
It is an object of the invention to provide a kind of method and device of prediction video-see duration, being capable of Accurate Prediction user
Viewing duration to certain video, to provide foundation as each service provider's optimized allocation of resources.
The present invention is achieved in that a kind of method of prediction video-see duration, and methods described includes：
If detecting the predetermined registration operation of triggering prediction video-see duration, the identification information that current video watches user is obtained
With the attribute information of currently playing video；Wherein, the attribute information includes identification information and type information；
When obtaining preliminary viewing according to the attribute information of the identification information and the video of current video viewing user
Long predictive value；
According to the default correspondence between the type information and error transfer factor ratio of the type information and video of the video
Relation, obtains the corresponding error transfer factor ratio of the video；
Final viewing duration prediction value is calculated according to the preliminary viewing duration prediction value and the error transfer factor ratio.
On the other hand, the present invention provides a kind of device of prediction video-see duration, and described device includes：
Information acquisition unit, if the predetermined registration operation for detecting triggering prediction video-see duration, obtains current video
The attribute information of the identification information and currently playing video of viewing user；Wherein, the attribute information include identification information and
Type information；
Initial predicted value acquiring unit, for the identification information and the video of user are watched according to the current video
Attribute information obtains preliminary viewing duration prediction value；
Error transfer factor ratio acquisition unit, for the type information and mistake of type information and video according to the video
Default corresponding relation between difference adjustment ratio, obtains the corresponding error transfer factor ratio of the video；
Final predictor calculation unit, based on according to the preliminary viewing duration prediction value and the error transfer factor ratio
Calculate final viewing duration prediction value.
If the present invention predicts the predetermined registration operation of video-see duration by detecting triggering, current video viewing user is obtained
Identification information and currently playing video attribute information；Wherein, the attribute information includes identification information and type information；
The attribute information that the identification information and the video of user are watched according to the current video obtains preliminary viewing duration prediction value；
According to the default corresponding relation between the type information and error transfer factor ratio of the type information and video of the video, obtain
The corresponding error transfer factor ratio of the video；Calculated most according to the preliminary viewing duration prediction value and the error transfer factor ratio
Duration prediction value is watched eventually such that it is able to viewing duration of the Accurate Prediction user to certain video, think each service provider
Optimized allocation of resources provides foundation.
Description of the drawings
Fig. 1 is a kind of schematic flow diagram of the method for prediction video-see duration provided in an embodiment of the present invention；
Fig. 2 is a kind of schematic flow diagram of the method for prediction video-see duration that another embodiment of the present invention is provided；
Fig. 3 is a kind of schematic block diagram of the device of prediction video-see duration provided in an embodiment of the present invention；
Fig. 4 is a kind of schematic block diagram of the device of prediction video-see duration that another embodiment of the present invention is provided.
Specific embodiment
In order that the objects, technical solutions and advantages of the present invention become more apparent, it is below in conjunction with drawings and Examples, right
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, and
It is not used in the restriction present invention.
Fig. 1 is referred to, Fig. 1 is a kind of exemplary flow of the method for prediction video-see duration provided in an embodiment of the present invention
Figure.In the present embodiment, the executive agent of the method for prediction video-see duration is video playing terminal.Video playing terminal can
Think the terminals such as PC, mobile phone and intelligent television, or other-end, it is not limited herein.The present embodiment
In prediction video-see duration method may comprise steps of：
S101：If detecting the predetermined registration operation of triggering prediction video-see duration, the mark that current video watches user is obtained
The attribute information of knowledge information and currently playing video.Wherein, the attribute information includes identification information and type information.
If video playing terminal detects the predetermined registration operation of triggering prediction video-see duration, current video viewing is obtained
The attribute information of the identification information of user and currently playing video.
Wherein, predict that the predetermined registration operation of video-see duration can be determined according to the actual requirements, be not limited herein.
For example, predict that the predetermined registration operation of video-see duration can play video to click on, even detect user's click and play a certain
Video, then it is assumed that trigger the predetermined registration operation of prediction video-see duration, video playing terminal is played by the video of program request, and is obtained
Take the attribute information of the identification information and currently playing video of current video viewing user.Certainly, predict video-see duration
Predetermined registration operation can also be currently playing video arrange interface performed by setting operation.Even detect user to exist
Click on after playing video, corresponding setting is performed at the setting interface of the video and operated, then it is assumed that trigger prediction video observing
The predetermined registration operation of duration is seen, video playing terminal obtains the identification information and currently playing video of current video viewing user
Attribute information.
The identification information of current video viewing user is used for the identity of the user for identifying the currently playing video of viewing.Currently
The identification information of video-see user can be the account information of the current user for logging in video playing terminal, or it is current
Video playing terminal sequence number information, be configured with specific reference to actual demand, be not limited herein.
The attribute information of video includes the corresponding identification information of video and type information.
Wherein, the identification information of video is used for the identity information for identifying video.For example, the identification information of video can be regarding
Frequently corresponding ID (Identity, proof of identification) number.It is configured with specific reference to actual demand, is not limited herein.
The type information of video is used to identify the type belonging to video.For example, type information can include：Variety show,
Movie or television play etc..More specifically, the type information of video can include：City play, love story, documentary film or educational film
Deng.It is configured with specific reference to practical situation, is not limited herein.
S102：The attribute information that the identification information and the video of user are watched according to the current video obtains preliminary sight
See duration prediction value.
Attribute of the video playing terminal in the identification information and currently playing video for getting current video viewing user
After information, when obtaining preliminary viewing according to the attribute information of the identification information and currently playing video of current video viewing user
Long predictive value.
Wherein, preliminary viewing duration prediction value is used for preliminary forecasting of the mark to the viewing duration of currently playing video
Value, i.e., tentatively watch current video viewing viewing of the user to currently playing video that duration prediction value refers specifically to preliminary forecasting
Duration.Preliminary viewing duration prediction value may exist certain and the actual final viewing duration value of currently playing video between
Error.
The preliminary historical viewing data or current video for watching the video that duration prediction value can be currently playing according to is seen
See the calculated viewing duration prediction of historical viewing data of the video that user couple belongs to same type with currently playing video
Value.And the corresponding relation between the identification information of video and preliminary viewing duration prediction value can be stored in advance in video playback end
In the data base at end.
Wherein, the historical viewing data of currently playing video includes corresponding at least one history of currently playing video
Viewing duration, i.e., each self-corresponding viewing duration when being watched by different user including currently playing video, or including currently broadcasting
When the video put is watched multiple by same user, each corresponding viewing duration.
The historical viewing data of the video that current video viewing user couple belongs to same type with currently playing video includes
The user corresponding at least one history viewing duration when watching to the video of this type.
Video playing terminal can be according to the corresponding relation between the identification information of video and preliminary viewing duration prediction value
Obtain the preliminary viewing duration prediction value of currently playing video.
S103：According to default between the type information and error transfer factor ratio of the type information and video of the video
Corresponding relation, obtains the corresponding error transfer factor ratio of the video.
Type information and error transfer factor of the video playing terminal according to the type information and video of currently playing video
Default corresponding relation between ratio, obtains the currently playing corresponding error transfer factor ratio of video.
Wherein, error transfer factor ratio is used to identify the error between preliminary viewing duration prediction value and actual viewing duration value
Adjustment ratio.Error transfer factor ratio for carrying out error transfer factor to preliminary viewing duration prediction value, to reduce currently playing regarding
Error between the preliminary viewing duration prediction value of frequency and actual viewing duration value.
The error transfer factor ratio of video can be obtained to the historical viewing data of all types of videos according to all users.
Each type of video corresponds to an error transfer factor ratio.Formed between the type information and error transfer factor ratio of video default
Corresponding relation.
Default corresponding relation between the type information and error transfer factor ratio of video can be stored in advance in video playback
In the data base of terminal.It is determined with specific reference to actual demand, is not limited herein.
S104：Final viewing duration prediction is calculated according to the preliminary viewing duration prediction value and the error transfer factor ratio
Value.
Video playing terminal is getting the corresponding preliminary viewing duration prediction value of currently playing video and error transfer factor
After ratio, calculate currently playing according to the currently playing corresponding preliminary viewing duration prediction value of video and error transfer factor ratio
Video it is corresponding it is final viewing duration prediction value.
Wherein, final viewing duration prediction value is used for final prediction of the mark to the viewing duration of currently playing video
Value.By this it is final viewing duration prediction value come characterize current video watch user currently playing video may be watched when
It is long.
Specifically, video playing terminal can pass through the ratio for calculating preliminary viewing duration prediction value and error transfer factor ratio
Value, is finally watched duration prediction value.
If above as can be seen that a kind of method of prediction video-see duration of the present embodiment offer is by detecting triggering
The predetermined registration operation of prediction video-see duration, obtains the category of the identification information and currently playing video of current video viewing user
Property information；Wherein, the attribute information includes identification information and type information；The mark of user is watched according to the current video
The attribute information of information and the video obtains preliminary viewing duration prediction value；According to the type information and video of the video
Type information and error transfer factor ratio between default corresponding relation, obtain the corresponding error transfer factor ratio of the video；Root
Final viewing duration prediction value is calculated according to the preliminary viewing duration prediction value and the error transfer factor ratio such that it is able to accurately
Prediction viewing duration of the user to certain video, to provide foundation as each service provider's optimized allocation of resources.
Fig. 2 is referred to, Fig. 2 is a kind of signal of the method for prediction video-see duration that another embodiment of the present invention is provided
Flow chart.In the present embodiment, the executive agent of the method for prediction video-see duration is video playing terminal.Video playback end
End can be the terminals such as PC, mobile phone and intelligent television, or other-end, it is not limited herein.This reality
The method for applying the prediction video-see duration in example may comprise steps of：
S201：If detecting the predetermined registration operation of triggering prediction video-see duration, the mark that current video watches user is obtained
The attribute information of knowledge information and currently playing video；Wherein, the attribute information includes identification information and type information.
The realization of step S101 in the specific implementation of step S201 in the present embodiment embodiment corresponding with Fig. 1
Mode is identical, specifically refers to the associated description of step S101 in the corresponding embodiments of Fig. 1, and here is omitted.
S202：According to the identification information of the video, obtain calculated according to the historical viewing data of the video
First viewing duration prediction value.
Attribute of the video playing terminal in the identification information and currently playing video for getting current video viewing user
After information, according to the identification information of currently playing video, obtain and calculated according to the historical viewing data of currently playing video
Arrive first viewing duration prediction value.
Wherein, the historical viewing data of currently playing video includes corresponding at least one history of currently playing video
Viewing duration, i.e., each self-corresponding viewing duration when being watched by different user including currently playing video, or including currently broadcasting
When the video put is watched multiple by same user, each corresponding viewing duration.
First viewing duration prediction value is used to identify the average viewing duration of currently playing video.
Further, before step S201, predict that the method for video-see duration may comprise steps of：
The viewing duration of each video is counted with the timing of the first predetermined period；Wherein, the viewing duration of each video
Including at least one viewing duration；
According to the viewing duration of each video, the average viewing duration of each video is calculated；
The identification information of associated storage each video and averagely watch duration.
Wherein, the first predetermined period can be configured according to practical situation, be not limited herein.For example, first preset
Cycle can be the viewing duration that a day, i.e. video playing terminal regularly count each video daily.
Timing counts corresponding timing and can be configured according to the actual requirements, for example, can be united with every every other hour
Meter once, is not limited herein.
The viewing duration of each video refers specifically to each video and is clicked on after broadcasting, when secondary playing duration by user.
The viewing duration of each video includes the corresponding at least one viewing duration of the video, i.e., including each video by not
Each self-corresponding viewing duration when watching with user, or when being watched multiple by same user including each video, it is corresponding every time
Viewing duration.
After video playing terminal counts on the corresponding at least one viewing duration of each video, according to each video correspondence
At least one viewing duration, calculate the corresponding average viewing duration of each video.Specifically, the corresponding average sight of each video
See that duration can be averaged by corresponding to each video at least one viewing duration to obtain.
Video playing terminal after the corresponding average viewing duration of each video is calculated, the mark of associated storage each video
Knowledge information and averagely watch duration.
S203：The type information of the identification information and the video of user is watched according to the current video, basis is obtained
The calculated second viewing duration prediction value of historical viewing data of the current video viewing user.
Attribute of the video playing terminal in the identification information and currently playing video for getting current video viewing user
After information, according to the type information of the identification information and currently playing video of the user, the conception of history according to the user is obtained
See the calculated second viewing duration prediction value of data.
Wherein, the historical viewing data of current video viewing user is referred specifically to the user couple and is belonged to currently playing video
The historical viewing data of the video of same type, its when can watch to the video of this type including the user it is corresponding extremely
Few history watches duration.
Second viewing duration prediction value is used to identify current video viewing average viewing of the user to currently playing video
Duration.
Video playing terminal can according to current video watch user historical viewing data calculate the user couple with it is current
The video of broadcasting belongs to the average viewing duration of the video of same type, and the user couple is belonged to same type with currently playing video
Video average viewing duration of the average viewing duration as the user to currently playing video.
Further, before step S201, predict that the method for video-see duration may comprise steps of：
Viewing duration of each user to each type of video is counted with the timing of the second predetermined period；Wherein, it is described every
Individual user includes at least one viewing duration to the viewing duration of each type of video；
According to viewing duration of each user to each type of video, each user is calculated to each type of video
Average viewing duration；
Average viewing duration of the identification information and each user of associated storage each user to each type of video.
Wherein, the second predetermined period can be configured according to practical situation, be not limited herein.For example, second preset
Cycle can be the viewing duration that a day, i.e. video playing terminal regularly count each video daily.
Timing counts corresponding timing and can be configured according to the actual requirements, for example, can be united with every every other hour
Meter once, is not limited herein.
To the viewing duration of each type of video, each user includes that each user is seen to each type of video
Corresponding at least one viewing duration when seeing.When being that each user is watched at least one times to each type of video, every time
Corresponding viewing duration.
Video playing terminal can obtain the identification information of user, and seeing when user's click broadcasting video is detected
After the completion of seeing, statistics obtains the viewing duration that the user clicks on the video of broadcasting to which.
Video playing terminal can count at least one viewing of each user to each type of video according to the method
Duration, and after at least one viewing duration of each user to each type of video is counted on, according to each user to every
At least one viewing duration of the video of type, calculates average viewing duration of each user to each type of video.Tool
Body, each user to the average viewing duration of each type of video can by each user to each type of video
At least one viewing duration average and obtain.
Video playing terminal is associated after each user average viewing duration corresponding to each type of video is calculated
Store the identification information and each user average viewing duration corresponding to each type of video of each user.
S204：The meansigma methodss of the first viewing duration prediction value and the second viewing duration prediction value are calculated, is obtained
The preliminary viewing duration prediction value.
Video playing terminal is seen to first after the first viewing duration prediction value and the second viewing duration prediction value is got
See that duration prediction value and the second viewing duration prediction value are averaged, that is, obtain tentatively watching duration prediction value.
Wherein, preliminary viewing duration prediction value is used for preliminary forecasting of the mark to the viewing duration of currently playing video
Value, i.e., tentatively watch viewing duration of the user to currently playing video that duration prediction value refers specifically to preliminary forecasting.It is preliminary to see
See and there is certain error between duration prediction value and the actual viewing duration value of currently playing video.
S205：According to default between the type information and error transfer factor ratio of the type information and video of the video
Corresponding relation, obtains the corresponding error transfer factor ratio of the video.
Type information and error transfer factor of the video playing terminal according to the type information and video of currently playing video
Default corresponding relation between ratio, obtains the currently playing corresponding error transfer factor ratio of video.
Wherein, error transfer factor ratio is used to identify the error between preliminary viewing duration prediction value and actual viewing duration value
Adjustment ratio.Error transfer factor ratio for carrying out error transfer factor to preliminary viewing duration prediction value, to reduce currently playing regarding
Error between the preliminary viewing duration prediction value of frequency and actual viewing duration value.
The error transfer factor ratio of video can be obtained to the historical viewing data of all types of videos according to all users.
Each type of video corresponds to an error transfer factor ratio.Formed between the type information and error transfer factor ratio of video default
Corresponding relation.
Further, before step S201, predict that the method for video-see duration may comprise steps of：
Training decision-making regression tree model；
Determined according to the decision-making regression tree model default between the type information of the video and error transfer factor ratio
Corresponding relation.
Video playing terminal can be with the 3rd predetermined period timing training determining based on the historical viewing data of total user
Plan regression tree model.
Wherein, decision-making regression tree model is specifically for exporting corresponding with the video of the type according to the type information of video
Error transfer factor ratio.When decision-making regression tree model is trained, the input value of decision-making regression tree model includes：Each type of video
Corresponding type information, duration information, total playback volume information, reproduction time segment information and user are to during the viewing of the type video
Long accounting information etc.；The output valve (returning object) of decision-making regression tree model is the corresponding error transfer factor of each type of video
Ratio.
3rd predetermined period can be configured according to practical situation, be not limited herein.For example, the 3rd predetermined period can
Think that one day, i.e. video playing terminal regularly count the viewing duration of each video daily.
Timing counts corresponding timing and can be configured according to the actual requirements, for example, can be united with every every other hour
Meter once, is not limited herein.
Video playing terminal determines the type of video according to decision-making regression tree model after decision-making regression tree model is trained
Default corresponding relation between information and error transfer factor ratio.And will be pre- between the type information of video and error transfer factor ratio
If corresponding relation is stored.
S206：Final viewing duration prediction is calculated according to the preliminary viewing duration prediction value and the error transfer factor ratio
Value.
The realization of step S104 in the specific implementation of step S206 in the present embodiment embodiment corresponding with Fig. 1
Mode is identical, specifically refers to the associated description of step S104 in the corresponding embodiments of Fig. 1, and here is omitted.
If above as can be seen that a kind of method of prediction video-see duration of the present embodiment offer is by detecting triggering
The predetermined registration operation of prediction video-see duration, obtains the category of the identification information and currently playing video of current video viewing user
Property information；Wherein, the attribute information includes identification information and type information；The mark of user is watched according to the current video
The attribute information of information and the video obtains preliminary viewing duration prediction value；According to the type information and video of the video
Type information and error transfer factor ratio between default corresponding relation, obtain the corresponding error transfer factor ratio of the video；Root
Final viewing duration prediction value is calculated according to the preliminary viewing duration prediction value and the error transfer factor ratio such that it is able to accurately
Prediction viewing duration of the user to certain video, to provide foundation as each service provider's optimized allocation of resources.
Fig. 3 is referred to, Fig. 3 shows a kind of signal of the device of prediction video-see duration provided in an embodiment of the present invention
Property block diagram.The device of the prediction video-see duration in the present embodiment is video playing terminal.Video playing terminal can be individual
The terminals such as people's computer, mobile phone and intelligent television, or other-end, it is not limited herein.In video playing terminal
Each unit be used to perform each step in the corresponding embodiments of Fig. 1, specifically refer in the corresponding embodiments of Fig. 1 and Fig. 1
Associated description, do not repeat herein.The device of the prediction video-see duration in the present embodiment can include：Information acquisition unit
301st, initial predicted value acquiring unit 302, error transfer factor ratio acquisition unit 303 and final predictor calculation unit 304.
If information acquisition unit 301 is used for the predetermined registration operation for detecting triggering prediction video-see duration, forward sight is worked as in acquisition
The attribute information of the identification information and currently playing video of frequency viewing user；Wherein, the attribute information includes identification information
And type information.Information acquisition unit 301 is watched the attribute of the identification information and currently playing video of user by current video
Information is sent to initial predicted value acquiring unit 302 and error transfer factor ratio acquisition unit 303.
Initial predicted value acquiring unit 302 is used for the current video viewing user's that receive information acquiring unit 301 sends
The attribute information of identification information and currently playing video, watches the identification information of user according to the current video and described regards
The attribute information of frequency obtains preliminary viewing duration prediction value.Initial predicted value acquiring unit 302 will tentatively watch duration prediction value
Send to final predictor calculation unit 304.
Error transfer factor ratio acquisition unit 303 is used for the current video viewing user that receive information acquiring unit 301 sends
Identification information and currently playing video attribute information, according to the type information of the type information and video of the video
Default corresponding relation between error transfer factor ratio, obtains the corresponding error transfer factor ratio of the video.Error transfer factor ratio
Acquiring unit 303 sends error transfer factor ratio to final predictor calculation unit 304.
Final predictor calculation unit 304 is used to receive the preliminary viewing duration of the transmission of initial predicted value acquiring unit 302
The error transfer factor ratio that predictive value and error transfer factor ratio acquisition unit 303 send, according to the preliminary viewing duration prediction value
Final viewing duration prediction value is calculated with the error transfer factor ratio.
If above as can be seen that a kind of device of prediction video-see duration of the present embodiment offer is by detecting triggering
The predetermined registration operation of prediction video-see duration, obtains the category of the identification information and currently playing video of current video viewing user
Property information；Wherein, the attribute information includes identification information and type information；The mark of user is watched according to the current video
The attribute information of information and the video obtains preliminary viewing duration prediction value；According to the type information and video of the video
Type information and error transfer factor ratio between default corresponding relation, obtain the corresponding error transfer factor ratio of the video；Root
Final viewing duration prediction value is calculated according to the preliminary viewing duration prediction value and the error transfer factor ratio such that it is able to accurately
Prediction viewing duration of the user to certain video, to provide foundation as each service provider's optimized allocation of resources.
Fig. 4 is referred to, Fig. 4 shows a kind of device of prediction video-see duration that another embodiment of the present invention is provided
Schematic block diagram.The device of the prediction video-see duration in the present embodiment is video playing terminal.Video playing terminal can be with
For terminals such as PC, mobile phone and intelligent televisions, or other-end, it is not limited herein.Video playback end
Each unit in end is used to perform each step in the corresponding embodiments of Fig. 2, specifically refers to the corresponding enforcements of Fig. 2 and Fig. 2
Associated description in example, is not repeated herein.The device of the prediction video-see duration in the present embodiment can include：Acquisition of information
Unit 401, initial predicted value acquiring unit 402, error transfer factor ratio acquisition unit 403 and final predictor calculation unit 404.
Wherein, initial predicted value acquiring unit 402 includes：First predictive value acquiring unit 420, the second predictive value obtain single
Unit 421 and initial predicted value computing unit 422.
If information acquisition unit 401 is used for the predetermined registration operation for detecting triggering prediction video-see duration, forward sight is worked as in acquisition
The attribute information of the identification information and currently playing video of frequency viewing user；Wherein, the attribute information includes identification information
And type information.Information acquisition unit 401 is watched the attribute of the identification information and currently playing video of user by current video
Information is sent to initial predicted value acquiring unit 402 and error transfer factor ratio acquisition unit 403.
The first predictive value acquiring unit 420 in initial predicted value acquiring unit 402 is used for receive information acquiring unit 401
The identification information and the attribute information of currently playing video of the current video viewing user of transmission, according to the mark of the video
Information, obtains the calculated first viewing duration prediction value of historical viewing data according to the video；Wherein, described first
Viewing duration prediction value is used for the average viewing duration for identifying the video.First predictive value acquiring unit 420 is watched first
Duration prediction value is sent to initial predicted value computing unit 422.
The second predictive value acquiring unit 421 in initial predicted value acquiring unit 402 is used for receive information acquiring unit 401
The identification information and the attribute information of currently playing video of the current video viewing user of transmission, sees according to the current video
The type information of the identification information and the video of user is seen, the history viewing number that user is watched according to the current video is obtained
According to the calculated second viewing duration prediction value；Wherein, the second viewing duration prediction value is used to identifying and described works as forward sight
Frequency viewing average viewing duration of the user to the video.Second predictive value acquiring unit 421 watches duration prediction value by second
Send to initial predicted value computing unit 422.
Initial predicted value computing unit 422 is used to receive the first viewing duration that the first predictive value acquiring unit 420 sends
The second viewing duration prediction value that predictive value and the second predictive value acquiring unit 421 send, calculates the first viewing duration pre-
The meansigma methodss of measured value and the second viewing duration prediction value, obtain the preliminary viewing duration prediction value.Initial predicted value meter
Calculation unit 422 will tentatively be watched duration prediction value and send to final predictor calculation unit 404.
Error transfer factor ratio acquisition unit 403 is used for the current video viewing user that receive information acquiring unit 401 sends
Identification information and currently playing video attribute information, according to the type information of the type information and video of the video
Default corresponding relation between error transfer factor ratio, obtains the corresponding error transfer factor ratio of the video.Error transfer factor ratio
Acquiring unit 403 sends error transfer factor ratio to final predictor calculation unit 404.
Final predictor calculation unit 404 is used to receive the preliminary viewing duration of the transmission of initial predicted value computing unit 422
The error transfer factor ratio that predictive value and error transfer factor ratio acquisition unit 403 send, according to the preliminary viewing duration prediction value
Final viewing duration prediction value is calculated with the error transfer factor ratio.
Further, predict that the device of video-see duration also includes：First statistic unit, the first average calculation unit
And first memory element.
First statistic unit is for counting the viewing duration of each video with the timing of the first predetermined period；Wherein, it is described every
The viewing duration of individual video includes at least one viewing duration.First statistic unit sends the viewing duration of each video to
One average calculation unit.
First average calculation unit is used for the viewing duration for receiving each video that the first statistic unit sends, according to institute
The viewing duration of each video is stated, the average viewing duration of each video is calculated.First average calculation unit is by each video
Average viewing duration send to the first memory element.
First memory element is used for the average viewing duration for receiving each video that the first average calculation unit sends, and closes
Connection stores the identification information of each video and averagely watches duration.
Further, predict that the device of video-see duration also includes：Second statistic unit, the second average calculation unit
And second memory element.
Second statistic unit for the second predetermined period timing count viewing of each user to each type of video
Duration；Wherein, described each user includes at least one viewing duration to the viewing duration of each type of video.Second statistics
Each user is sent to the second average calculation unit by unit to the viewing duration of each type of video.
Second average calculation unit is used for each user for receiving the transmission of the second statistic unit to each type of video
Viewing duration, according to viewing duration of each user to each type of video, calculate each user and regard to each type of
The average viewing duration of frequency.Second average calculation unit is by each user to long hair during the average viewing of each type of video
Deliver to the second memory element.
Second memory element is used for each user for receiving the transmission of the second average calculation unit to each type of video
Average viewing duration, the average viewing of the identification information and each user of associated storage each user to each type of video
Duration.
Further, predict that the device of video-see duration also includes：Training unit and determining unit.
Training unit is used to train decision-making regression tree model；Wherein, the input value of the decision-making regression tree model includes：Often
The corresponding type information of video of type, duration information, total playback volume information, reproduction time segment information and user are to the type
The viewing duration accounting information of video；The output valve of the decision-making regression tree model is that the corresponding error of each type of video is adjusted
Whole ratio；The error transfer factor ratio is used to identify the mistake between the preliminary viewing duration prediction value and actual viewing duration value
Difference adjustment ratio.Training unit sends decision-making regression tree model to determining unit.
Determining unit is used for the decision-making regression tree model for receiving training unit transmission, true according to the decision-making regression tree model
Default corresponding relation between the type information and error transfer factor ratio of the fixed video.
If above as can be seen that a kind of device of prediction video-see duration of the present embodiment offer is by detecting triggering
The predetermined registration operation of prediction video-see duration, obtains the category of the identification information and currently playing video of current video viewing user
Property information；Wherein, the attribute information includes identification information and type information；The mark of user is watched according to the current video
The attribute information of information and the video obtains preliminary viewing duration prediction value；According to the type information and video of the video
Type information and error transfer factor ratio between default corresponding relation, obtain the corresponding error transfer factor ratio of the video；Root
Final viewing duration prediction value is calculated according to the preliminary viewing duration prediction value and the error transfer factor ratio such that it is able to accurately
Prediction viewing duration of the user to certain video, to provide foundation as each service provider's optimized allocation of resources.
One of ordinary skill in the art will appreciate that：The step of realizing said method embodiment or part steps can pass through
Completing, aforesaid program can be stored in computer read/write memory medium the related hardware of programmed instruction, and the program exists
During execution, the step of including said method embodiment is performed, and aforesaid storage medium includes：ROM, RAM, magnetic disc or CD
Etc. it is various can be with the medium of store program codes.
Presently preferred embodiments of the present invention is the foregoing is only, not to limit the present invention, all essences in the present invention
Any modification, equivalent and improvement made within god and principle etc., should be included within the scope of the present invention.
Claims (10)
1. it is a kind of prediction video-see duration method, it is characterised in that methods described includes：
If detecting the predetermined registration operation of triggering prediction video-see duration, obtain the identification information of current video viewing user and work as
The attribute information of the video of front broadcasting；Wherein, the attribute information includes identification information and type information；
The preliminary viewing duration of attribute information acquisition that the identification information and the video of user are watched according to the current video is pre-
Measured value；
According to the default corresponding relation between the type information and error transfer factor ratio of the type information and video of the video,
Obtain the corresponding error transfer factor ratio of the video；
Final viewing duration prediction value is calculated according to the preliminary viewing duration prediction value and the error transfer factor ratio.
2. the method for predicting video-see duration as claimed in claim 1, it is characterised in that described according to the current video
The attribute information of the identification information and the video of viewing user obtains preliminary viewing duration prediction value to be included：
According to the identification information of the video, when acquisition is watched according to the historical viewing data calculated first of the video
Long predictive value；Wherein, the first viewing duration prediction value is used to identify the average viewing duration of the video；
The type information of the identification information and the video of user is watched according to the current video, acquisition works as forward sight according to described
The calculated second viewing duration prediction value of historical viewing data of frequency viewing user；Wherein, the second viewing duration is pre-
Measured value is used to identify the current video viewing average viewing duration of the user to the video；
The meansigma methodss of the first viewing duration prediction value and the second viewing duration prediction value are calculated, the preliminary sight is obtained
See duration prediction value.
3. the method for predicting video-see duration as claimed in claim 2, it is characterised in that if described detect triggering prediction
The predetermined registration operation of video-see duration, obtains the attribute letter of the identification information and currently playing video of current video viewing user
Before breath, methods described also includes：
The viewing duration of each video is counted with the timing of the first predetermined period；Wherein, the viewing duration of each video includes
At least one viewing duration；
According to the viewing duration of each video, the average viewing duration of each video is calculated；
The identification information of associated storage each video and averagely watch duration.
4. the method for predicting video-see duration as claimed in claim 2, it is characterised in that if described detect triggering prediction
The predetermined registration operation of video-see duration, obtains the attribute letter of the identification information and currently playing video of current video viewing user
Before breath, methods described also includes：
Viewing duration of each user to each type of video is counted with the timing of the second predetermined period；Wherein, described each use
Family includes at least one viewing duration to the viewing duration of each type of video；
According to viewing duration of each user to each type of video, each user is calculated to the average of each type of video
Viewing duration；
Average viewing duration of the identification information and each user of associated storage each user to each type of video.
5. the method for predicting video-see duration as claimed in claim 2, it is characterised in that if described detect triggering prediction
The predetermined registration operation of video-see duration, obtains the attribute letter of the identification information and currently playing video of current video viewing user
Before breath, methods described also includes：
Training decision-making regression tree model；Wherein, the input value of the decision-making regression tree model includes：Each type of video correspondence
Type information, duration information, total playback volume information, reproduction time segment information and user accounted for the viewing duration of the type video
Compare information；The output valve of the decision-making regression tree model is the corresponding error transfer factor ratio of each type of video；The error
Adjustment ratio is used to identify the error transfer factor ratio between the preliminary viewing duration prediction value and actual viewing duration value；
The default correspondence between the type information of the video and error transfer factor ratio is determined according to the decision-making regression tree model
Relation.
6. it is a kind of prediction video-see duration device, it is characterised in that described device includes：
Information acquisition unit, if the predetermined registration operation for detecting triggering prediction video-see duration, obtains current video viewing
The attribute information of the identification information of user and currently playing video；Wherein, the attribute information includes identification information and type
Information；
Initial predicted value acquiring unit, for the attribute of the identification information and the video of user is watched according to the current video
Acquisition of information tentatively watches duration prediction value；
Error transfer factor ratio acquisition unit, the type information and error for type information and video according to the video are adjusted
Default corresponding relation between whole ratio, obtains the corresponding error transfer factor ratio of the video；
Final predictor calculation unit, for being calculated most according to the preliminary viewing duration prediction value and the error transfer factor ratio
Duration prediction value is watched eventually.
7. the device of video-see duration is predicted as claimed in claim 6, it is characterised in that the initial predicted value obtains single
Unit includes：
First predictive value acquiring unit, for the identification information according to the video, obtains and is watched according to the history of the video
The calculated first viewing duration prediction value of data；Wherein, the first viewing duration prediction value is used to identify the video
Average viewing duration；
Second predictive value acquiring unit, for the type of the identification information and the video of user is watched according to the current video
Information, obtains the calculated second viewing duration prediction value of historical viewing data for watching user according to the current video；
Wherein, the second viewing duration prediction value is used to identifying the current video viewing user to during the average viewing of the video
It is long；
Initial predicted value computing unit, for calculating the first viewing duration prediction value and the second viewing duration prediction value
Meansigma methodss, obtain it is described it is preliminary viewing duration prediction value.
8. the device of video-see duration is predicted as claimed in claim 7, it is characterised in that described device also includes：
First statistic unit, for the viewing duration of each video is counted with the timing of the first predetermined period；Wherein, it is described each regard
The viewing duration of frequency includes at least one viewing duration；
First average calculation unit, for the viewing duration according to each video, calculates the average viewing of each video
Duration；
First memory element, identification information for associated storage each video and averagely watches duration.
9. the device of video-see duration is predicted as claimed in claim 7, it is characterised in that described device also includes：
Second statistic unit, during for counting each user to the viewing of each type of video with the timing of the second predetermined period
It is long；Wherein, described each user includes at least one viewing duration to the viewing duration of each type of video；
Second average calculation unit, for the viewing duration according to each user to each type of video, calculates each use
Average viewing duration of the family to each type of video；
Second memory element, the identification information and each user for associated storage each user are put down to each type of video
Duration is watched.
10. the device of video-see duration is predicted as claimed in claim 7, it is characterised in that described device also includes：
Training unit, for training decision-making regression tree model；Wherein, the input value of the decision-making regression tree model includes：It is every kind of
The corresponding type information of video of type, duration information, total playback volume information, reproduction time segment information and user are regarded to the type
The viewing duration accounting information of frequency；The output valve of the decision-making regression tree model is the corresponding error transfer factor of each type of video
Ratio；The error transfer factor ratio is used to identify the error between the preliminary viewing duration prediction value and actual viewing duration value
Adjustment ratio；
Determining unit, for determined according to the decision-making regression tree model video type information and error transfer factor ratio it
Between default corresponding relation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611249506.1A CN106604137B (en) | 2016-12-29 | 2016-12-29 | Method and device for predicting video watching duration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611249506.1A CN106604137B (en) | 2016-12-29 | 2016-12-29 | Method and device for predicting video watching duration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106604137A true CN106604137A (en) | 2017-04-26 |
| CN106604137B CN106604137B (en) | 2020-06-12 |
Family
ID=58604239
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611249506.1A Active CN106604137B (en) | 2016-12-29 | 2016-12-29 | Method and device for predicting video watching duration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106604137B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107864234A (en) * | 2017-12-18 | 2018-03-30 | 广东省电信规划设计院有限公司 | The acquisition methods and device of address renewed treaty time |
| CN108632670A (en) * | 2018-03-15 | 2018-10-09 | 北京奇艺世纪科技有限公司 | A kind of video satisfaction determines method and device |
| CN110149540A (en) * | 2018-04-27 | 2019-08-20 | 腾讯科技（深圳）有限公司 | Recommendation process method, apparatus, terminal and the readable medium of multimedia resource |
| CN110491419A (en) * | 2019-08-26 | 2019-11-22 | 广东小天才科技有限公司 | A kind of control method for playing back, system and terminal device |
| CN110933492A (en) * | 2019-12-10 | 2020-03-27 | 北京爱奇艺科技有限公司 | Method and device for predicting playing time |
| CN113132803A (en) * | 2021-04-23 | 2021-07-16 | Oppo广东移动通信有限公司 | Video watching time length prediction method, device, storage medium and terminal |
| CN113496422A (en) * | 2021-09-07 | 2021-10-12 | 济宁景泽信息科技有限公司 | Block chain-based popularization resource allocation method and big data information cloud platform |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1777279A (en) * | 2005-12-22 | 2006-05-24 | 李欣 | Method and system for automatically selecting programmes for user |
| JP2008294839A (en) * | 2007-05-25 | 2008-12-04 | Mitsubishi Electric Corp | Digital broadcast receiving system |
| EP2352292A1 (en) * | 2009-12-07 | 2011-08-03 | Telefonaktiebolaget L M Ericsson (PUBL) | System for managing television channels |
| CN102802056A (en) * | 2012-09-12 | 2012-11-28 | 北京播思软件技术有限公司 | Method used for inserting advertisement in digital broadcasting television program |
| CN103338403A (en) * | 2012-09-17 | 2013-10-02 | 中国传媒大学 | Broadcasting television system and personalized program recommending method in system |
| CN103440335A (en) * | 2013-09-06 | 2013-12-11 | 北京奇虎科技有限公司 | Video recommendation method and device |
| CN104462573A (en) * | 2014-12-29 | 2015-03-25 | 北京奇艺世纪科技有限公司 | Method and device for displaying video retrieval results |
| CN104584001A (en) * | 2012-08-22 | 2015-04-29 | 兰屈克有限公司 | Systems and methods for projecting viewership data |
| CN104936023A (en) * | 2015-06-11 | 2015-09-23 | 嘉兴市广播电视集团 | Big data collecting and analyzing method and system of digital television user behavior |
| CN105163142A (en) * | 2015-09-09 | 2015-12-16 | 青岛海信传媒网络技术有限公司 | User preference determination method, video recommendation method, user preference determination system and video recommendation system |
| CN105701226A (en) * | 2016-01-18 | 2016-06-22 | 合网络技术（北京）有限公司 | Multimedia resource assessment method and device |
| CN106028156A (en) * | 2016-06-24 | 2016-10-12 | 合肥工业大学 | Television viewer interest modeling method and system |
-
2016
- 2016-12-29 CN CN201611249506.1A patent/CN106604137B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1777279A (en) * | 2005-12-22 | 2006-05-24 | 李欣 | Method and system for automatically selecting programmes for user |
| JP2008294839A (en) * | 2007-05-25 | 2008-12-04 | Mitsubishi Electric Corp | Digital broadcast receiving system |
| EP2352292A1 (en) * | 2009-12-07 | 2011-08-03 | Telefonaktiebolaget L M Ericsson (PUBL) | System for managing television channels |
| CN104584001A (en) * | 2012-08-22 | 2015-04-29 | 兰屈克有限公司 | Systems and methods for projecting viewership data |
| CN102802056A (en) * | 2012-09-12 | 2012-11-28 | 北京播思软件技术有限公司 | Method used for inserting advertisement in digital broadcasting television program |
| CN103338403A (en) * | 2012-09-17 | 2013-10-02 | 中国传媒大学 | Broadcasting television system and personalized program recommending method in system |
| CN103440335A (en) * | 2013-09-06 | 2013-12-11 | 北京奇虎科技有限公司 | Video recommendation method and device |
| CN104462573A (en) * | 2014-12-29 | 2015-03-25 | 北京奇艺世纪科技有限公司 | Method and device for displaying video retrieval results |
| CN104936023A (en) * | 2015-06-11 | 2015-09-23 | 嘉兴市广播电视集团 | Big data collecting and analyzing method and system of digital television user behavior |
| CN105163142A (en) * | 2015-09-09 | 2015-12-16 | 青岛海信传媒网络技术有限公司 | User preference determination method, video recommendation method, user preference determination system and video recommendation system |
| CN105701226A (en) * | 2016-01-18 | 2016-06-22 | 合网络技术（北京）有限公司 | Multimedia resource assessment method and device |
| CN106028156A (en) * | 2016-06-24 | 2016-10-12 | 合肥工业大学 | Television viewer interest modeling method and system |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107864234B (en) * | 2017-12-18 | 2020-12-11 | 广东省电信规划设计院有限公司 | Method and device for acquiring address continuation time |
| CN107864234A (en) * | 2017-12-18 | 2018-03-30 | 广东省电信规划设计院有限公司 | The acquisition methods and device of address renewed treaty time |
| CN108632670A (en) * | 2018-03-15 | 2018-10-09 | 北京奇艺世纪科技有限公司 | A kind of video satisfaction determines method and device |
| CN110149540A (en) * | 2018-04-27 | 2019-08-20 | 腾讯科技（深圳）有限公司 | Recommendation process method, apparatus, terminal and the readable medium of multimedia resource |
| CN110149540B (en) * | 2018-04-27 | 2021-08-24 | 腾讯科技（深圳）有限公司 | Recommendation processing method and device for multimedia resources, terminal and readable medium |
| CN110491419A (en) * | 2019-08-26 | 2019-11-22 | 广东小天才科技有限公司 | A kind of control method for playing back, system and terminal device |
| CN110491419B (en) * | 2019-08-26 | 2021-08-13 | 广东小天才科技有限公司 | Playing control method, system and terminal equipment |
| CN110933492B (en) * | 2019-12-10 | 2022-03-04 | 北京爱奇艺科技有限公司 | Method and device for predicting playing time |
| CN110933492A (en) * | 2019-12-10 | 2020-03-27 | 北京爱奇艺科技有限公司 | Method and device for predicting playing time |
| CN113132803A (en) * | 2021-04-23 | 2021-07-16 | Oppo广东移动通信有限公司 | Video watching time length prediction method, device, storage medium and terminal |
| CN113132803B (en) * | 2021-04-23 | 2022-09-16 | Oppo广东移动通信有限公司 | Video watching time length prediction method, device, storage medium and terminal |
| CN113496422B (en) * | 2021-09-07 | 2021-12-03 | 济宁景泽信息科技有限公司 | Block chain-based popularization resource allocation method and big data information cloud platform |
| CN113496422A (en) * | 2021-09-07 | 2021-10-12 | 济宁景泽信息科技有限公司 | Block chain-based popularization resource allocation method and big data information cloud platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106604137B (en) | 2020-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106604137A (en) | Method and apparatus for predicting video viewing time length | |
| US11568431B2 (en) | Methods and apparatus to compensate for server-generated errors in database proprietor impression data due to misattribution and/or non-coverage | |
| US11349943B2 (en) | Methods and apparatus for adjusting model threshold levels | |
| US11765061B2 (en) | Methods and apparatus to predict end of streaming media using a prediction model | |
| US9277265B2 (en) | Methods and apparatus to calculate video-on-demand and dynamically inserted advertisement viewing probability | |
| US11068928B2 (en) | Methods and apparatus to determine impressions corresponding to market segments | |
| JP2020526088A (en) | Method, apparatus, computer program and system for determining information about a viewer of an audiovisual content program | |
| CN103686236A (en) | Method and system for recommending video resource | |
| WO2013112911A1 (en) | Systems, methods, and articles of manufacture to measure online audiences | |
| USRE49120E1 (en) | Methods and apparatus to determine a duration of media presentation based on tuning session duration | |
| CN113763027B (en) | Recommendation information processing method, recommendation information generation method and device | |
| CN112291625B (en) | Information quality processing method, information quality processing device, electronic equipment and storage medium | |
| CN104506892A (en) | Data adjustment method and device | |
| AU2016213749A1 (en) | Methods and apparatus to characterize households with media meter data |
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 | ||
| CB02 | Change of applicant information |
Address after: 516006 TCL technology building, No.17, Huifeng Third Road, Zhongkai high tech Zone, Huizhou City, Guangdong Province Applicant after: TCL Technology Group Co., Ltd Address before: 516006 Guangdong province Huizhou Zhongkai hi tech Development Zone No. nineteen District Applicant before: TCL RESEARCH AMERICA Inc. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |