CN107948752A - Subscribe to main broadcaster's sort method, device and terminal - Google Patents
Subscribe to main broadcaster's sort method, device and terminal Download PDFInfo
- Publication number
- CN107948752A CN107948752A CN201711120851.XA CN201711120851A CN107948752A CN 107948752 A CN107948752 A CN 107948752A CN 201711120851 A CN201711120851 A CN 201711120851A CN 107948752 A CN107948752 A CN 107948752A
- Authority
- CN
- China
- Prior art keywords
- main broadcaster
- user
- subscription
- days
- weight coefficient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4667—Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/251—Learning process for intelligent management, e.g. learning user preferences for recommending movies
-
- 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]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
- H04N21/25866—Management of end-user data
- H04N21/25891—Management of end-user data being end-user 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
- H04N21/43—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
- H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Social Psychology (AREA)
- Computer Graphics (AREA)
- Computing Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of subscription main broadcaster's sort method, including:Obtain user behavior data of the user in live platform to subscription main broadcaster;User behavior feature is obtained according to the user behavior data;The user behavior feature includes at least two characteristic indexs;The characteristic index is pressed into normalized, obtains normalization characteristic desired value;Normalization characteristic desired value is multiplied by this feature index and presets weight coefficient in timeliness, obtains weighting normalization characteristic index value;The weighting normalization characteristic index value of various features index is summed, obtains the score value of the subscription main broadcaster;According to the score value of each subscription main broadcaster, it is ranked up to subscribing to main broadcaster.The subscription main broadcaster sequence of the present invention preferably can match the favorable rating for subscribing to main broadcaster with user, and then user is had the live enthusiasm of the viewing of higher and more preferable product usage experience.Present invention also offers one kind to subscribe to main broadcaster's collator and terminal.
Description
Technical field
The present invention relates to order models technical field, and specifically, the present invention relates to one kind to subscribe to main broadcaster's sort method, dress
Put and terminal.
Background technology
Network direct broadcasting possesses huge main broadcaster colony and viewing as a kind of emerging social media of current Internet era
Live user group.User can subscribe to the main broadcaster oneself liked, the subscription master that system would generally be to user on live platform
Broadcast and be ranked up, and the subscription main broadcaster of former of ranking is showed in the outermost layer that the user subscribes to main broadcaster's entrance.
In the prior art, usually to subscribe to main broadcaster, the online order of number from big to small arranges each subscription main broadcaster at the same time
Sequence.However, the subscription main broadcaster of the ranking several former obtained by the sort method is frequently not the favorite subscription main broadcaster of user,
User needs more to be operated, and can just find and oneself be most liked into more pages from outermost subscription main broadcaster entrance
Subscription main broadcaster, reduce user and watch live enthusiasm and the satisfaction of Product Experience.
The content of the invention
The purpose of the present invention is intended at least solve one of above-mentioned technological deficiency, and particularly existing subscription main broadcaster sequence is not
Can be by the favorite technological deficiency subscribed to main broadcaster and ranked the first of user.
The present invention provides a kind of subscription main broadcaster's sort method, includes the following steps:
Obtain user behavior data of the user in live platform to subscription main broadcaster in predetermined period;
User behavior feature is obtained according to the user behavior data;The user behavior feature includes at least two features
Index;
The characteristic index is pressed into normalized, obtains normalization characteristic desired value;Normalization characteristic desired value is multiplied
The weight coefficient in timeliness is preset with this feature index, obtains weighting normalization characteristic index value;By adding for various features index
The summation of normalization characteristic desired value is weighed, obtains the score value of the subscription main broadcaster;
According to the score value of each subscription main broadcaster, it is ranked up to subscribing to main broadcaster.
Preferably, the characteristic index includes:
Subscribe to apart from number of days and payment amount;
It is described to subscribe to the present number of days of time gap for subscribing to current subscription main broadcaster in live platform for user apart from number of days;
Payment amount of the payment amount for user to current subscription main broadcaster;
It is described subscribe to apart from number of days it is corresponding normalization index be:1- (subscription is apart from number of days/predetermined period);
The corresponding normalization index of the payment amount is:Payment amount/the user is in total payment gold of live platform
Volume.
Further, the characteristic index further includes:
Whether whether newest viewing, viewing duration, viewing distance number of days, viewing number of days, bullet screen number, payment number of days and pay
Take;
It is described that whether newest viewing watches current subscription master for user's last day when the viewing of live platform is live
Broadcast live;
A length of user watches the live duration of current subscription main broadcaster during the viewing;
The viewing distance number of days is the present day of the live time gap of user last time viewing current subscription main broadcaster
Number;
The viewing number of days has the live number of days of viewing current subscription main broadcaster for user;
The barrage number sends the number of barrage for user during viewing current subscription main broadcaster is live;
The payment number of days has the number of days paid to current subscription main broadcaster for user;
It is described whether to pay as whether user pays current subscription main broadcaster.
Preferably, the score value according to each subscription main broadcaster, after being ranked up to subscription main broadcaster, further includes:
According to sequence from big to small, pushed to user and subscribe to main broadcaster.
Preferably, the sequence of the basis from big to small, pushes to user and subscribes to main broadcaster, including:
According to sequence from big to small, to the subscription main broadcaster come above of user push preset quantity;Or
According to sequence from big to small, using the subscription main broadcaster come above of push preset quantity as outer anchor,
The remaining subscription main broadcaster of user's push pushes outer anchor and internal layer main broadcaster as internal layer main broadcaster to the user;Wherein,
The outer anchor is showed in the user and is showed in subscription main broadcaster's overview window of live platform, the internal layer main broadcaster from institute
State and subscribe to subscription main broadcaster's overview page that main broadcaster's overview window enters.
Preferably, the sequence of the basis from big to small, pushes after subscribing to main broadcaster to user, further includes:
Statistics preset duration user clicks on the probability of the subscription main broadcaster to rank the first;
The weight coefficient of characteristic index is adjusted according to the probability;
The statistics preset duration user clicks on the probability of the subscription main broadcaster to rank the first, including:
Obtain ranking the first to user's push for whole server in preset duration and subscribe to number of users and the user of main broadcaster
Feedback clicks on the number of users that main broadcaster is subscribed in the push that ranks the first, and user feedback is clicked on to the use that main broadcaster is subscribed in the push that ranks the first
Amount amount divided by the number of users for the subscription main broadcaster that ranks the first to user's push, obtain probability;Or
Obtain the number of users that the subscription main broadcaster that ranks the first is pushed to user that colony is extracted in preset duration and user is anti-
Feed point hits the number of users that main broadcaster is subscribed in the push that ranks the first, and user feedback is clicked on to the user that main broadcaster is subscribed in the push that ranks the first
Quantity divided by the number of users for the subscription main broadcaster that ranks the first to user's push, obtain probability;
The weight coefficient that characteristic index is adjusted according to the probability, including:
Confirm that the probability is less than preset value, the weight coefficient of each characteristic index is changed to another default weight system
Number;Or
Confirm that the probability is less than preset value, the weight coefficient of each characteristic index added to the adjusted value of this feature index,
The adjusted value be on the occasion of or negative value.
Preferably, after the weight coefficient that characteristic index is adjusted according to the probability, further include:
The probability of different preset durations is compared, using the weight coefficient of the preset duration of maximum probability as it is default when
Weight coefficient in effect;Wherein, the weight coefficient of different preset durations is different.
Preferably, it is described to obtain user behavior data of the user in live platform to each subscription main broadcaster, bag in predetermined period
Include:
Whole server whole user is randomly divided into the user group of default quantity;Wherein, the feature of each user group refers to
Weight coefficient in the default timeliness of mark is different;
Obtain user behavior data of the user of each user group in predetermined period in live platform to each subscription main broadcaster;
The score value according to each subscription main broadcaster, after being ranked up to subscription main broadcaster, further includes:
According to sequence from big to small, pushed to user and subscribe to main broadcaster;
Count the probability that each user group preset duration user clicks on the subscription main broadcaster to rank the first;
The weight coefficient of the characteristic index of the user group of maximum probability is obtained, using the weight coefficient as whole server
Weight coefficient in default timeliness.
The present invention also provides one kind to subscribe to main broadcaster's collator, it includes:
User behavior data acquisition module:For obtaining user behavior data of the user in live platform to subscription main broadcaster;
User behavior feature acquisition module:For obtaining user behavior feature according to the user behavior data;The use
Family behavioural characteristic includes at least two characteristic indexs;
Subscribe to main broadcaster's score value computing module:For the characteristic index to be pressed normalized, obtain normalization characteristic and refer to
Scale value;Normalization characteristic desired value is multiplied by the weight coefficient that this feature index presets timeliness, weighting normalization feature is obtained and refers to
Scale value;The weighting normalization characteristic index value of various features index is summed, obtains the score value of the subscription main broadcaster;
Subscribe to main broadcaster's sorting module:For the score value according to each subscription main broadcaster, it is ranked up to subscribing to main broadcaster.
The present invention also provides a kind of terminal, it includes:
One or more processors;
Memory;
One or more application program, wherein one or more of application programs are stored in the memory and quilt
It is configured to be performed by one or more of processors, one or more of programs are configured to:It is preferred to perform any of the above
Subscription main broadcaster's sort method described in embodiment.
Subscription main broadcaster sort method, device and terminal provided by the invention, by analyzing the user in the user preset cycle
Behavioral data obtains user behavior feature, and according to every characteristic index that can effectively reflect user preferences in user behavior feature
Like model to build the subscription main broadcaster of user, the subscription main broadcaster sequence obtained by subscription main broadcaster hobby model can be preferably
Match with user to the favorable rating for subscribing to main broadcaster, and then user is had the live enthusiasm of the viewing of higher and more preferable product
Usage experience.
Secondly, subscription main broadcaster sort method provided by the invention to subscribing to main broadcaster also by periodically liking each feature in model
The weight coefficient of index is adjusted, and is obtained each characteristic index and is preset weight coefficient in timeliness, it is ensured that subscribes to main broadcaster's sequence
As a result timeliness and accuracy, further such that arrive subscribe to main broadcaster sequence can preferably with user to subscribe to main broadcaster happiness
Love degree matches, and then user is had the live enthusiasm of the viewing of higher and more preferable product usage experience.
The additional aspect of the present invention and advantage will be set forth in part in the description, these will become from the following description
Obtain substantially, or recognized by the practice of the present invention.
Brief description of the drawings
Of the invention above-mentioned and/or additional aspect and advantage will become from the following description of the accompanying drawings of embodiments
Substantially and it is readily appreciated that, wherein:
Fig. 1 is subscription main broadcaster's sort method flow chart of first embodiment of the invention;
Fig. 2 is the index definition list of one embodiment of the invention;
Fig. 3 is subscription main broadcaster's sort method flow chart of second embodiment of the invention;
Fig. 4 is subscription main broadcaster's sort method flow chart of third embodiment of the invention;
Fig. 5 is subscription main broadcaster's collator schematic diagram of one embodiment of the invention;
Fig. 6 is the terminal structure schematic diagram of the present invention.
Embodiment
The embodiment of the present invention is described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning to end
Same or similar label represents same or similar element or has the function of same or like element.Below with reference to attached
The embodiment of figure description is exemplary, and is only used for explaining the present invention, and is not construed as limiting the claims.
The present invention provides one kind to subscribe to main broadcaster's sort method, as shown in Figure 1, the subscription master for first embodiment of the invention
Sort method flow chart is broadcast, is included the following steps:
Step S101:Obtain user behavior data of the user in live platform to subscription main broadcaster in predetermined period.
Specifically, in embodiments of the present invention, user of the user in live platform to subscription main broadcaster in nearest 30 days is obtained
Behavioral data.Wherein, the main broadcaster oneself liked that main broadcaster subscribes to for user on live platform, above-mentioned user behavior data are subscribed to
The action and corresponding data that user respectively performs each subscription main broadcaster in be that live plateform system is recorded nearest 30 days.
The predetermined period is used to limit the time interval that acquired user behavior data occurs, so that data have timeliness
Property and easy to by the cycle timing update and management.Need it is clear that, the predetermined period and from the limitation of the present embodiment,
The predetermined period can be one week, one month, the time cycle such as 45 days, and those skilled in the art can be according to practical application need
Ask and the predetermined period is adjusted.
Step S102:User behavior feature is obtained according to the user behavior data;The user behavior feature includes extremely
Few two characteristic indexs.
Specifically, user divides the user behavior data for subscribing to main broadcaster in nearest 30 days obtained to step S101
Analysis, excavation obtain with the corresponding user behavior feature of user behavior data, wherein, comprising at least in the user behavior feature
Two characteristic index.In embodiments of the present invention, the user behavior feature obtained includes following nine characteristic indexs:
Subscribe to apart from number of days, payment amount, whether it is newest viewing, viewing duration, viewing distance number of days, viewing number of days, bullet
Whether curtain number, payment number of days and pay.
The definition of above-mentioned nine characteristic indexs is specific as follows:
It is that user is present in the time gap of live platform subscription current subscription main broadcaster in nearest 30 days to subscribe to apart from number of days
Number of days;
Payment amount of the payment amount for user in nearest 30 days to current subscription main broadcaster;
Whether newest viewing whether watched when the viewing of live platform is live for user's last day in nearest 30 days it is current
It is live to subscribe to main broadcaster;
User watches the live duration of current subscription main broadcaster in a length of nearest 30 days during viewing;
Viewing distance number of days is that the time gap that user's last time viewing current subscription main broadcaster is live in nearest 30 days shows
Number of days;
Viewing number of days is that user has the live number of days of viewing current subscription main broadcaster in nearest 30 days;
Barrage number is the number that user sends barrage during viewing current subscription main broadcaster is live in nearest 30 days;
Payment number of days is that user has the number of days paid to current subscription main broadcaster in nearest 30 days;
Whether payment is that whether user pays current subscription main broadcaster in nearest 30 days.
The characteristic index can embody user to it is each subscription main broadcaster favorable rating, in other embodiments, from
The quantity for the characteristic index that family behavioral data excavates can be increased or decreased according to actual conditions are all, characteristic index
Definition be also not limited to nine above-mentioned characteristic indexs, can also be other equally can feedback user to subscribe to main broadcaster like journey
The validity feature index of degree.
Step S103:The characteristic index is pressed into normalized, obtains normalization characteristic desired value;By normalization characteristic
Desired value, which is multiplied by this feature index, presets weight coefficient in timeliness, obtains weighting normalization characteristic index value;By various features
The weighting normalization characteristic index value summation of index, obtains the score value of the subscription main broadcaster.
Specifically, as shown in Fig. 2, for the corresponding normalization characteristic index of above-mentioned each characteristic index in the embodiment of the present invention and
It, which is normalized, defines weight coefficient corresponding with each characteristic index in nearly 30 days.In the present embodiment, each characteristic index is pressed it
Normalization index as shown in Figure 2 is normalized, and obtains normalization characteristic desired value;Each normalization characteristic is referred to again
Scale value be multiplied by with the weight coefficient in the corresponding default timeliness of characteristic index, wherein, in embodiments of the present invention, weight coefficient
Default timeliness be 30 days, the weight coefficient ratio of each characteristic index meets the following conditions:Subscribe to apart from number of days: payment gold
Volume: whether newest viewing: viewing duration: viewing distance number of days: viewing number of days: bullet screen number: payment number of days: whether pay=3:
2.5∶2∶1.5∶1.5∶1.2∶1∶0.8∶0.5;Weighting normalizing is obtained according to normalization characteristic desired value and corresponding weight coefficient
Change characteristic index value, then each weighting normalization characteristic index value is summed, obtain that happiness of the user to subscription main broadcaster can be embodied
The score value of love degree.
The default timeliness is used for the usage time section for limiting the weight coefficient of each characteristic index, so that weight coefficient has
Time-effectiveness and easy to by the cycle timing update and management.Need it is clear that, the default timeliness and from the present embodiment
Limitation, the default timeliness is generally longer than or equal to the predetermined period in step S101, can be one week, one month,
The time span such as 45 days, those skilled in the art can be adjusted the default timeliness according to practical application request.
Step S104:According to the score value of each subscription main broadcaster, it is ranked up to subscribing to main broadcaster.
Specifically, each score value for subscribing to main broadcaster obtained by step S103 is integrated, and is arranged by score value subscribing to main broadcaster
Sequence.In embodiments of the present invention, the order according to score value from big to small, is ranked up, i.e. the big subscription of score value to subscribing to main broadcaster
Main broadcaster stands out.
Subscription main broadcaster's sort method of the embodiment of the present invention, is obtained by analyzing the user behavior data in the user preset cycle
User is built to user behavior feature, and according to every characteristic index that can effectively reflect user preferences in user behavior feature
Subscription main broadcaster like model, by subscription main broadcaster like model obtain subscription main broadcaster sequence can be preferably with user to ordering
The favorable rating for readding main broadcaster matches, and then user is had the live enthusiasm of the viewing of higher and more preferable product usage experience.
As shown in figure 3, being subscription main broadcaster's sort method flow chart of second embodiment of the invention, what which provided orders
Main broadcaster's sort method is read, is included the following steps:
Step S201:Obtain user behavior data of the user in live platform to subscription main broadcaster in predetermined period.
Specifically, in embodiments of the present invention, user of the user in live platform to subscription main broadcaster in nearest 30 days is obtained
Behavioral data.Wherein, the main broadcaster oneself liked that main broadcaster subscribes to for user on live platform, above-mentioned user behavior data are subscribed to
The action and corresponding data that user respectively performs each subscription main broadcaster in be that live plateform system is recorded nearest 30 days.
The predetermined period is used to limit the time interval that acquired user behavior data occurs, so that data have timeliness
Property and easy to by the cycle timing update and management.Need it is clear that, the predetermined period and from the limitation of the present embodiment,
The predetermined period can be one week, one month, the time cycle such as 45 days, and those skilled in the art can be according to practical application need
Ask and the predetermined period is adjusted.
Step S202:User behavior feature is obtained according to the user behavior data;The user behavior feature includes extremely
Few two characteristic indexs.
Specifically, user divides the user behavior data for subscribing to main broadcaster in nearest 30 days obtained to step S201
Analysis, excavation obtain with the corresponding user behavior feature of user behavior data, wherein, comprising at least in the user behavior feature
Two characteristic index.In embodiments of the present invention, the user behavior feature obtained includes following nine characteristic indexs:
Subscribe to apart from number of days, payment amount, whether it is newest viewing, viewing duration, viewing distance number of days, viewing number of days, bullet
Whether curtain number, payment number of days and pay.
The definition of above-mentioned nine characteristic indexs is specific as follows:
It is that user is present in the time gap of live platform subscription current subscription main broadcaster in nearest 30 days to subscribe to apart from number of days
Number of days;
Payment amount of the payment amount for user in nearest 30 days to current subscription main broadcaster;
Whether newest viewing whether watched when the viewing of live platform is live for user's last day in nearest 30 days it is current
It is live to subscribe to main broadcaster;
User watches the live duration of current subscription main broadcaster in a length of nearest 30 days during viewing;
Viewing distance number of days is that the time gap that user's last time viewing current subscription main broadcaster is live in nearest 30 days shows
Number of days;
Viewing number of days is that user has the live number of days of viewing current subscription main broadcaster in nearest 30 days;
Barrage number is the number that user sends barrage during viewing current subscription main broadcaster is live in nearest 30 days;
Payment number of days is that user has the number of days paid to current subscription main broadcaster in nearest 30 days;
Whether payment is that whether user pays current subscription main broadcaster in nearest 30 days.
The characteristic index can embody user to it is each subscription main broadcaster favorable rating, in other embodiments, from
The quantity for the characteristic index that family behavioral data excavates can be increased or decreased according to actual conditions are all, characteristic index
Definition be also not limited to nine above-mentioned characteristic indexs, can also be other equally can feedback user to subscribe to main broadcaster like journey
The validity feature index of degree.
Step S203:The characteristic index is pressed into normalized, obtains normalization characteristic desired value;By normalization characteristic
Desired value, which is multiplied by this feature index, presets weight coefficient in timeliness, obtains weighting normalization characteristic index value;By various features
The weighting normalization characteristic index value summation of index, obtains the score value of the subscription main broadcaster.
Specifically, as shown in Fig. 2, for the corresponding normalization characteristic index of above-mentioned each characteristic index in the embodiment of the present invention and
It, which is normalized, defines weight coefficient corresponding with each characteristic index in nearly 30 days.In the present embodiment, each characteristic index is pressed it
Normalization index as shown in Figure 2 is normalized, and obtains normalization characteristic desired value;Each normalization characteristic is referred to again
Scale value be multiplied by with the weight coefficient in the corresponding default timeliness of characteristic index, wherein, in embodiments of the present invention, weight coefficient
Default timeliness be 30 days, the weight coefficient ratio of each characteristic index meets the following conditions:Subscribe to apart from number of days: payment gold
Volume: whether newest viewing: viewing duration: viewing distance number of days: viewing number of days: bullet screen number: payment number of days: whether pay=3:
2.5∶2∶1.5∶1.5∶1.2∶1∶0.8∶0.5;Weighting normalizing is obtained according to normalization characteristic desired value and corresponding weight coefficient
Change characteristic index value, then each weighting normalization characteristic index value is summed, obtain that happiness of the user to subscription main broadcaster can be embodied
The score value of love degree.
The default timeliness is used for the usage time section for limiting the weight coefficient of each characteristic index, so that weight coefficient has
Time-effectiveness and easy to by the cycle timing update and management.Need it is clear that, the default timeliness and from the present embodiment
Limitation, the default timeliness is generally longer than or equal to the predetermined period in step S201, can be one week, one month,
The time span such as 45 days, those skilled in the art can be adjusted the default timeliness according to practical application request.
Step S204:According to the score value of each subscription main broadcaster, it is ranked up to subscribing to main broadcaster.
Specifically, each score value for subscribing to main broadcaster obtained by step S203 is integrated, and is arranged by score value subscribing to main broadcaster
Sequence.In embodiments of the present invention, the order according to score value from big to small, is ranked up, i.e. the big subscription of score value to subscribing to main broadcaster
Main broadcaster stands out.
Step S205:According to sequence from big to small, pushed to user and subscribe to main broadcaster.
Specifically, the sequence according to score value from big to small obtained according to step S204, each subscription main broadcaster is pushed away in order
Send to user in subscription main broadcaster's displayed page of live platform.
On live platform, the displayed page often more than one of the subscription main broadcaster of user.User is in of live platform
It is usually set to subscribe to the overview window of main broadcaster in people's data homepage, subscription main broadcaster entrance of the overview window as user, leads to
Often only have several main broadcasters to show position, user can directly select that a certain to be showed in subscription main broadcaster general by subscription main broadcaster entrance
The subscription main broadcaster of window is look at, to realize the homepage or direct broadcasting room that quickly and easily enter subscription main broadcaster;User also can by into
Single stepping, selects to enter and subscribes to main broadcaster's overview page, ordering for overview window is not shown in find from subscription main broadcaster entrance
Read main broadcaster.
In one embodiment, according to sequence from big to small, to the user push subscription come above of 5
Main broadcaster.
In another embodiment, according to sequence from big to small, 5 of the push subscription main broadcasters come above are made
For outer anchor, assign the remaining subscription main broadcaster of user push as internal layer main broadcaster, to the user push outer anchor with
Internal layer main broadcaster;Wherein, the outer anchor is showed in the user in the above-mentioned subscription main broadcaster overview window of live platform, described
Internal layer main broadcaster is showed in the subscription main broadcaster's overview page for entering from above-mentioned subscription main broadcaster overview window and obtaining.
Need it is clear that, the present invention to user push come above subscription main broadcaster quantity and from above-mentioned reality
The limitation of mode is applied, the quantity pushed has been adjusted according to practical situations, itself and above-mentioned subscription main broadcaster overview window
Show that digit is corresponding, can be the preset quantities such as 3,7,10.
Step S206:Statistics preset duration user clicks on the probability of the subscription main broadcaster to rank the first.
Specifically, in one embodiment, pushing to user for whole server in nearest 30 days is obtained to rank the first
Subscribe to the number of users of main broadcaster and the number of users that main broadcaster is subscribed in the push that ranks the first is clicked in user feedback, user feedback is clicked on
The push that ranks the first, which subscribes to the number of users of main broadcaster divided by ranks the first to user's push, subscribes to the number of users of main broadcaster, obtains general
Rate.Click on the subscription main broadcaster's to rank the first by calculating user in preset duration using the user of whole server as object
Probability, more accesses click probability the most accurate.
In another embodiment, obtain ranking the first to user's push for extraction colony in nearest 30 days and subscribe to main broadcaster's
The number of users that main broadcaster is subscribed in the push that ranks the first is clicked in number of users and user feedback, and user feedback click is ranked the first and is pushed away
Send the number of users for subscribing to main broadcaster divided by rank the first to user's push and subscribe to the number of users of main broadcaster, obtain probability;For example, from
Sample 100 users in live platform, is obtained to this 100 user's push with subscription main broadcaster's sort method provided by the invention
The subscription main broadcaster to rank the first, and count and the number of users that main broadcaster is subscribed in the push that ranks the first is clicked in 30 days, which is removed
With 100, probability is obtained.User is selected by way of sampling and clicks on the subscription to rank the first to calculate user in preset duration
The probability of main broadcaster, can reduce operand, quickly obtain accurately clicking on probability.
Similarly, the preset duration is used for the frequency for limiting statistical probability, so that the probability of statistics is easy to determine by the cycle
Shi Gengxin and management.Need it is clear that, the preset duration and from the limitation of the present embodiment, the preset duration can be with
The predetermined period in step S201 is identical or different, and the default timeliness for being generally equal to or being shorter than in step S203,
To ensure that the weight coefficient during statistical probability does not change.The preset duration can be one week, one month, the time such as 45 days
Length, those skilled in the art can be adjusted the default timeliness according to practical application request.
Step S207:The weight coefficient of characteristic index is adjusted according to the probability.
Specifically, the probability of the subscription main broadcaster to rank the first is clicked on by counting preset duration user, can be learnt by this
Whether the subscription main broadcaster sequence that the weight coefficient of each characteristic index of proportioning is drawn feeds back user to subscribing to main broadcaster's exactly
Favorable rating.According to the probability, the weight coefficient of each characteristic index is adjusted, obtains the weight of new each characteristic index
Coefficient matches.
In one embodiment, when confirming that the probability is less than preset value, then by the weight coefficient of each characteristic index more
It is changed to another default weight coefficient;For example, the probability that the subscription main broadcaster that user's click ranks the first is obtained when statistics is less than
When 50%, the weight coefficient of each characteristic index is changed to following another default weight coefficient:
Subscribe to apart from number of days: payment amount: whether newest viewing: viewing duration: viewing distance number of days: viewing number of days: bullet
Screen number: payment number of days :=2: 3: 2.5: 1.5: 1.5: 1: 1.5: 1: 0.2 whether is paid.
In another embodiment, when confirming that the probability is less than preset value, the weight coefficient of each characteristic index is added
The adjusted value of upper this feature index, the adjusted value be on the occasion of or negative value;For example, click on what is ranked the first when statistics obtains user
When subscribing to the probability of main broadcaster less than 65%, the weight coefficient of each said features index is added into its corresponding adjusted value, it is each described
The weight coefficient ratio of characteristic index is as follows:
Subscribe to apart from number of days: payment amount: whether newest viewing: viewing duration: viewing distance number of days: viewing number of days: bullet
Shield number: payment number of days: whether=(3-0.4): (2.5+0.5): (2-0.5): (1.5+0.2): (1.5-0.3): (1.2+ if paying
0.2)∶(1+0.5)∶(0.8+0.2)∶(0.5-0.3)。
The probability of the subscription main broadcaster to rank the first is clicked on by counting preset duration user, and according to the probability to each spy
The weight coefficient of sign index is adjusted so that the subscription main broadcaster sequence obtained by the weight coefficient can preferably with user couple
The favorable rating for subscribing to main broadcaster matches, and then makes user have the live enthusiasm of the viewing of higher and more preferable product to use body
Test.
Step S208:The probability of different preset durations is compared, by the weight coefficient of the preset duration of maximum probability
As the weight coefficient in scheduled duration.
Specifically, the probability that the user of several preset durations is clicked on to the subscription main broadcaster to rank the first is compared, will
The probability for the preset duration corresponding to maximum weight coefficient as the weight coefficient in default timeliness, wherein, difference
The weight coefficient of preset duration is different;For example, 6 times carried out in nearly 180 days with 30 days for a cycle statistics user's point
The probability for hitting the subscription main broadcaster to rank the first is compared, will if the probability of nearly 30 days statistics is the maximum in 6 groups of probability
The corresponding weight coefficient used in main broadcaster's sort method of subscribing to subscribes to main broadcaster's sort method as the lower first quarter in the duration
Weight coefficient.By being compared to historical statistics probability, the weight of each characteristic index the most suitable can be rapidly drawn
Coefficient matches, and then can be quickly obtained and sort with user to the subscription main broadcaster that the favorable rating for subscribing to main broadcaster matches.
As shown in figure 4, being subscription main broadcaster's sort method flow chart of third embodiment of the invention, what which provided orders
Main broadcaster's sort method is read, is included the following steps:
Step S301:Whole server whole user is randomly divided into the user group of default quantity;Wherein, each user group
Characteristic index to preset weight coefficient in timeliness different.
Specifically, whole server whole user is randomly divided into the user group of default quantity;Wherein, after each user group
It is different to continue the weight coefficient that the characteristic index used when subscribing to main broadcaster and sorting is preset in timeliness;It is for example, whole server is complete
Portion user is randomly divided into 5 user groups, and the characteristic index of each user group is preset to be preset with and other users group not phase in timeliness
Same weight coefficient.
Step S302:Obtain user row of the user of each user group in predetermined period in live platform to each subscription main broadcaster
For data.
Specifically, in embodiments of the present invention, the user of each user group in nearest 30 days is obtained in live platform to each
Subscribe to the user behavior data of main broadcaster.Wherein, it is the main broadcaster oneself liked that user subscribes on live platform to subscribe to main broadcaster, on
The user for stating each user group in nearest 30 days that user behavior data is that live plateform system is recorded respectively leads each subscribe to
Broadcast the action of execution and corresponding data.
The predetermined period is used to limit the time interval that acquired user behavior data occurs, so that data have timeliness
Property and easy to by the cycle timing update and management.Need it is clear that, the predetermined period and from the limitation of the present embodiment,
The predetermined period can be one week, one month, the time cycle such as 45 days, and those skilled in the art can be according to practical application need
Ask and the predetermined period is adjusted.
Step S303:User behavior feature is obtained according to the user behavior data;The user behavior feature includes extremely
Few two characteristic indexs.
Step S304:The characteristic index is pressed into normalized, obtains normalization characteristic desired value;By normalization characteristic
Desired value, which is multiplied by this feature index, presets weight coefficient in timeliness, obtains weighting normalization characteristic index value;By various features
The weighting normalization characteristic index value summation of index, obtains the score value of the subscription main broadcaster.
Step S305:According to the score value of each subscription main broadcaster, it is ranked up to subscribing to main broadcaster.
Step S306:According to sequence from big to small, pushed to user and subscribe to main broadcaster.
Wherein, the step subscribed in main broadcaster's sort method that above-mentioned steps S303 to S306 is provided with second embodiment of the invention
Rapid S202 to S205 is consistent, and details are not described herein again.
Step S307:Count the probability that each user group preset duration user clicks on the subscription main broadcaster to rank the first.
Specifically, in units of respective user in each user group, statistics preset duration user inside the group, which clicks on, comes head
The probability of the subscription main broadcaster of position, details are not described herein again for the statistic processes of the probability, its process is carried with second embodiment of the invention
The step S206 subscribed in main broadcaster's sort method of confession is consistent.In this step, the number of the probability counted and the use
The default quantity of family group is consistent.
Step S308:The weight coefficient of the characteristic index of the user group of maximum probability is obtained, using the weight coefficient as whole
Weight coefficient in the default timeliness of a server.
Specifically, the multiple probability obtained in above-mentioned steps S307 are compared, learn the probability for maximum institute
Corresponding user group, so that the weight coefficient of the characteristic index of the user group is obtained, using the weight coefficient as whole server
Default timeliness in weight coefficient.
By being different user groups by all user's random divisions, and the characteristic index of different user group is using different
Weight coefficient, is just rapidly drawing the weight coefficient proportioning of each characteristic index the most suitable after a preset duration,
As the weight coefficient in the default timeliness of whole server so that the subscription main broadcaster sequence that whole server obtains can
Match well with user to the favorable rating for subscribing to main broadcaster, and then user is had the live enthusiasm of the viewing of higher and more preferable
Product usage experience.
As shown in figure 5, being subscription main broadcaster's collator schematic diagram of one embodiment of the invention, subscription main broadcaster, which sorts, to be filled
Put including user behavior data acquisition module 501, user behavior feature acquisition module 502, subscribe to main broadcaster's score value computing module 503
With subscription main broadcaster's sorting module 504.
User behavior data acquisition module 501:For obtaining user in predetermined period in live platform to subscribing to main broadcaster's
User behavior data;
User behavior feature acquisition module 502:For obtaining user behavior feature according to the user behavior data;It is described
User behavior feature includes at least two characteristic indexs;
Subscribe to main broadcaster's score value computing module 503:For the characteristic index to be pressed normalized, normalization characteristic is obtained
Desired value;Normalization characteristic desired value is multiplied by this feature index and presets weight coefficient in timeliness, obtains weighting normalization spy
Levy desired value;The weighting normalization characteristic index value of various features index is summed, obtains the score value of the subscription main broadcaster;
Subscribe to main broadcaster's sorting module 504:For the score value according to each subscription main broadcaster, it is ranked up to subscribing to main broadcaster.
Preferably, the characteristic index includes:
Subscribe to apart from number of days and payment amount;
It is described to subscribe to the present number of days of time gap for subscribing to current subscription main broadcaster in live platform for user apart from number of days;
Payment amount of the payment amount for family to current subscription main broadcaster;
It is described subscribe to apart from number of days it is corresponding normalization index be:1- (subscription is apart from number of days/predetermined period);
The corresponding normalization index of the payment amount is:Payment amount/the user is in total payment gold of live platform
Volume.
Further, the characteristic index further includes:
Whether whether newest viewing, viewing duration, viewing distance number of days, viewing number of days, barrage number, payment number of days and pay
Take;
It is described that whether newest viewing watches current subscription master for user's last day when the viewing of live platform is live
Broadcast live;
A length of user watches the live duration of current subscription main broadcaster during the viewing;
The viewing distance number of days is the present day of the live time gap of user last time viewing current subscription main broadcaster
Number;
The viewing number of days has the live number of days of viewing current subscription main broadcaster for user;
The bullet screen number sends the number of barrage for user during viewing current subscription main broadcaster is live;
The payment number of days has the number of days paid to current subscription main broadcaster for user;
It is described whether to pay as whether user pays current subscription main broadcaster.
Preferably, the subscription main broadcaster sorting module 504, which further includes, subscribes to main broadcaster's pushing module, described according to each subscription
The score value of main broadcaster, after being ranked up to subscription main broadcaster, is used for:
According to sequence from big to small, pushed to user and subscribe to main broadcaster.
Preferably, subscription main broadcaster's pushing module is specifically used for:
According to sequence from big to small, to the subscription main broadcaster come above of user push preset quantity;Or
According to sequence from big to small, using the subscription main broadcaster come above of push preset quantity as outer anchor,
The remaining subscription main broadcaster of user's push pushes outer anchor and internal layer main broadcaster as internal layer main broadcaster to the user;Wherein,
The outer anchor is showed in the user and is showed in subscription main broadcaster's overview window of live platform, the internal layer main broadcaster from institute
State and subscribe to main broadcaster's overview window into obtained subscription main broadcaster's overview page.
Preferably, the subscription main broadcaster sorting module 504 further includes weight adjustment module, in the basis from big to small
Sequence, pushes after subscribing to main broadcaster to user, is used for:
Statistics preset duration user clicks on the probability of the subscription main broadcaster to rank the first;
The weight coefficient of characteristic index is adjusted according to the probability.
Preferably, the weight adjustment module clicks on the subscription main broadcaster's to rank the first with regard to the statistics preset duration user
Probability, is specifically used for:
Obtain ranking the first to user's push for whole server in preset duration and subscribe to number of users and the user of main broadcaster
Feedback clicks on the number of users that main broadcaster is subscribed in the push that ranks the first, and user feedback is clicked on to the use that main broadcaster is subscribed in the push that ranks the first
Amount amount divided by the number of users for the subscription main broadcaster that ranks the first to user's push, obtain probability;Or
Obtain the number of users that the subscription main broadcaster that ranks the first is pushed to user that colony is extracted in preset duration and user is anti-
Feed point hits the number of users that main broadcaster is subscribed in the push that ranks the first, and user feedback is clicked on to the user that main broadcaster is subscribed in the push that ranks the first
Quantity divided by the number of users for the subscription main broadcaster that ranks the first to user's push, obtain probability;
The weight adjustment module is specifically used for regard to the weight coefficient that characteristic index is adjusted according to the probability:
Confirm that the probability is less than preset value, the weight coefficient of each characteristic index is changed to another default weight system
Number;Or
Confirm that the probability is less than preset value, the weight coefficient of each characteristic index added to the adjusted value of this feature index,
The adjusted value be on the occasion of or negative value;
The weight adjustment module is additionally operable to:
After the weight coefficient that characteristic index is adjusted according to the probability, the probability of different preset durations is carried out
Compare, using the weight coefficient of the preset duration of maximum probability as the weight coefficient in default timeliness;Wherein, different preset durations
Weight coefficient it is different.
Preferably, the data acquisition module 501 is additionally operable to:
Whole server whole user is randomly divided into the user group of default quantity;Wherein, the feature of each user group refers to
Target weight coefficient is different;
Obtain user behavior data of the user of each user group in predetermined period in live platform to each subscription main broadcaster;
The subscription main broadcaster sorting module 504, in the score value according to each subscription main broadcaster, is ranked up to subscribing to main broadcaster
Afterwards, it is additionally operable to:
According to sequence from big to small, pushed to user and subscribe to main broadcaster;
Count the probability that each user group preset duration user clicks on the subscription main broadcaster to rank the first;
The weight coefficient of the characteristic index of the user group of maximum probability is obtained, using the weight coefficient as whole server
Weight coefficient in default timeliness.
Subscription main broadcaster's collator of the embodiment of the present invention, is obtained by analyzing the user behavior data in the user preset cycle
User is built to user behavior feature, and according to every characteristic index that can effectively reflect user preferences in user behavior feature
Subscription main broadcaster like model, by subscription main broadcaster like model obtain subscription main broadcaster sequence can be preferably with user to ordering
The favorable rating for readding main broadcaster matches, and then user is had the live enthusiasm of the viewing of higher and more preferable product usage experience.
Fig. 6 show terminal structure schematic diagram provided by the invention.The terminal of the embodiment of the present invention, can include one or
The more than one processor 601 of person, further includes memory 602, WiFi (wireless fidelity, Wireless Fidelity) circuit
603rd, RF (Radio Frequency, radio frequency) circuit 604, voicefrequency circuit 605, sensor 606, output equipment 607, input are set
Standby 604, power supply 609, processor 601 is the control centre of terminal, utilizes various interfaces and connection above each several part.Ability
Field technique personnel are appreciated that the restriction of terminal structure shown in Fig. 6 not structure paired terminal, can include more more than illustrating
Or less component, either combine some components or different components arrangement.
WiFi circuitry 603 can provide WLAN or internet access to the user;It may include antenna, WiFi module
Deng.RF circuits 604 can receive and send messages, or in communication process signal reception and transmission;It may include antenna, at least one puts
Big device, tuner, one or more oscillators, coupler, duplexer etc..The voice data that voicefrequency circuit 605 will can receive
Electric signal is converted into, is transferred to loudspeaker, also the voice signal that microphone is collected voice data can be converted to, issue processor
601 processing;It can set loudspeaker, microphone, earphone interface etc..Sensor 606 can be used for sensing outer signals, and issue place
Reason device 601 is handled;It may include motion sensor, optical sensor etc..Output equipment 607 can be used for showing various signals;It can
For using LCD (Liquid Crystal Display, liquid crystal display), OLED (Organic Light-Emitting
Diode, Organic Light Emitting Diode) etc. form configure display panel.Input equipment 604 can be used for the letters such as input numeral and character
Breath;It can be physical button, contact panel etc..Power supply 609 can be that terminal each several part is powered, and pass through power-supply management system and processing
609 logical connection of device;It may include one or more direct current or AC power, charging system, power supply status indicator
Deng component.Memory 602 can be used for storage software program and module;It can be computer-readable recording medium, be specifically
Hard disk, flash memory etc..Processor is the control centre of terminal, by running or performing the software program being stored in memory 602
And/or module, and the data for being stored in memory 602 are called, perform terminal various functions, processing terminal data.
As one embodiment, terminal includes:One or more processors 601, memory 602, one or more application
Program, wherein one or more of application programs are stored in memory 602 and are configured as by one or more of
Processor 601 performs, and one or more of programs are configured to carry out first, second, and third embodiment institute of the invention described above
Subscription main broadcaster's sort method of offer.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing module, can also
That unit is individually physically present, can also two or more units be integrated in a module.Above-mentioned integrated mould
Block can both be realized in the form of hardware, can also be realized in the form of software function module.The integrated module is such as
Fruit is realized in the form of software function module and as independent production marketing or in use, can also be stored in a computer
In read/write memory medium.Storage medium mentioned above can be read-only storage, disk or CD etc..
The above is only some embodiments of the present invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (10)
1. one kind subscribes to main broadcaster's sort method, it is characterised in that including:
Obtain user behavior data of the user in live platform to subscription main broadcaster in predetermined period;
User behavior feature is obtained according to the user behavior data;The user behavior feature refers to comprising at least two features
Mark;
The characteristic index is pressed into normalized, obtains normalization characteristic desired value;Normalization characteristic desired value is multiplied by this
Characteristic index presets the weight coefficient in timeliness, obtains weighting normalization characteristic index value;The weighting of various features index is returned
One changes the summation of characteristic index value, obtains the score value of the subscription main broadcaster;
According to the score value of each subscription main broadcaster, it is ranked up to subscribing to main broadcaster.
2. subscription main broadcaster's sort method according to claim 1, it is characterised in that the characteristic index includes:
Subscribe to apart from number of days and payment amount;
It is described to subscribe to the present number of days of time gap for subscribing to current subscription main broadcaster in live platform for user apart from number of days;
Payment amount of the payment amount for user to current subscription main broadcaster;
It is described subscribe to apart from number of days it is corresponding normalization index be:1- (subscription is apart from number of days/predetermined period);
The corresponding normalization index of the payment amount is:Payment amount/the user is in total payment amount of live platform.
3. subscription main broadcaster's sort method according to claim 2, it is characterised in that the characteristic index further includes:
Whether whether newest viewing, viewing duration, viewing distance number of days, viewing number of days, bullet screen number, payment number of days and pay;
It is described whether newest viewing for user's last day, whether to watch current subscription main broadcaster straight when the viewing of live platform is live
Broadcast;
A length of user watches the live duration of current subscription main broadcaster during the viewing;
The viewing distance number of days is the present number of days of the live time gap of user last time viewing current subscription main broadcaster;
The viewing number of days has the live number of days of viewing current subscription main broadcaster for user;
The barrage number sends the number of barrage for user during viewing current subscription main broadcaster is live;
The payment number of days has the number of days paid to current subscription main broadcaster for user;
It is described whether to pay as whether user pays current subscription main broadcaster.
4. subscription main broadcaster's sort method according to claim 1, it is characterised in that point according to each subscription main broadcaster
Value, after being ranked up to subscription main broadcaster, further includes:
According to sequence from big to small, pushed to user and subscribe to main broadcaster.
5. subscription main broadcaster's sort method according to claim 4, it is characterised in that the sequence of the basis from big to small,
Pushed to user and subscribe to main broadcaster, including:
According to sequence from big to small, to the subscription main broadcaster come above of user push preset quantity;Or
According to sequence from big to small, using the subscription main broadcaster come above of push preset quantity as outer anchor, described in
The remaining subscription main broadcaster of user's push pushes outer anchor and internal layer main broadcaster as internal layer main broadcaster to the user;Wherein, it is described
Outer anchor is showed in the user in subscription main broadcaster's overview window of live platform, and the internal layer main broadcaster, which is showed in from described, to order
Read subscription main broadcaster's overview page of main broadcaster's overview window entrance.
6. subscription main broadcaster's sort method according to claim 4, it is characterised in that the sequence of the basis from big to small,
Push after subscribing to main broadcaster, further include to user:
Statistics preset duration user clicks on the probability of the subscription main broadcaster to rank the first;
The weight coefficient of characteristic index is adjusted according to the probability;
The statistics preset duration user clicks on the probability of the subscription main broadcaster to rank the first, including:
Obtain ranking the first to user's push for whole server in preset duration and subscribe to number of users and the user feedback of main broadcaster
The number of users that main broadcaster is subscribed in the push that ranks the first is clicked on, user feedback click is ranked the first and pushes the number of users for subscribing to main broadcaster
Amount divided by the number of users for the subscription main broadcaster that ranks the first to user's push, obtain probability;Or
Obtain the number of users and user feedback point for the subscription main broadcaster that ranks the first to user's push that colony is extracted in preset duration
The number of users that main broadcaster is subscribed in the push that ranks the first is hit, user feedback is clicked on to the number of users for the push subscription main broadcaster that ranks the first
Divided by rank the first to user's push and subscribe to the number of users of main broadcaster, obtain probability;
The weight coefficient that characteristic index is adjusted according to the probability, including:
Confirm that the probability is less than preset value, the weight coefficient of each characteristic index is changed to another default weight coefficient;Or
Confirm that the probability is less than preset value, the weight coefficient of each characteristic index is added to the adjusted value of this feature index, it is described
Adjusted value be on the occasion of or negative value.
7. subscription main broadcaster's sort method according to claim 6, it is characterised in that described that feature is adjusted according to the probability
After the weight coefficient of index, further include:
The probability of different preset durations is compared, using the weight coefficient of the preset duration of maximum probability as in default timeliness
Weight coefficient;Wherein, the weight coefficient of different preset durations is different.
8. subscription main broadcaster's sort method according to claim 1, it is characterised in that:
User behavior data of the user in live platform to each subscription main broadcaster in the acquisition predetermined period, including:
Whole server whole user is randomly divided into the user group of default quantity;Wherein, the characteristic index of each user group is pre-
If the weight coefficient in timeliness is different;
Obtain user behavior data of the user of each user group in predetermined period in live platform to each subscription main broadcaster;
The score value according to each subscription main broadcaster, after being ranked up to subscription main broadcaster, further includes:
According to sequence from big to small, pushed to user and subscribe to main broadcaster;
Count the probability that each user group preset duration user clicks on the subscription main broadcaster to rank the first;
The weight coefficient of the characteristic index of the user group of maximum probability is obtained, using the weight coefficient as the default of whole server
Weight coefficient in timeliness.
9. one kind subscribes to main broadcaster's collator, it is characterised in that including:
User behavior data acquisition module:For obtaining user behavior of the user in live platform to subscription main broadcaster in predetermined period
Data;
User behavior feature acquisition module:For obtaining user behavior feature according to the user behavior data;User's row
It is characterized and includes at least two characteristic indexs;
Subscribe to main broadcaster's score value computing module:For the characteristic index to be pressed normalized, normalization characteristic desired value is obtained;
Normalization characteristic desired value is multiplied by this feature index and presets weight coefficient in timeliness, obtains weighting normalization characteristic index
Value;The weighting normalization characteristic index value of various features index is summed, obtains the score value of the subscription main broadcaster;
Subscribe to main broadcaster's sorting module:For the score value according to each subscription main broadcaster, it is ranked up to subscribing to main broadcaster.
10. a kind of terminal, it is characterised in that it includes:
One or more processors;
Memory;
One or more application program, wherein one or more of application programs are stored in the memory and are configured
To be performed by one or more of processors, one or more of programs are configured to, and perform such as claim 1 to 8 times
Subscription main broadcaster's sort method described in one.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711120851.XA CN107948752B (en) | 2017-11-14 | 2017-11-14 | Ordering method, device and terminal for subscription anchor |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711120851.XA CN107948752B (en) | 2017-11-14 | 2017-11-14 | Ordering method, device and terminal for subscription anchor |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107948752A true CN107948752A (en) | 2018-04-20 |
CN107948752B CN107948752B (en) | 2021-01-08 |
Family
ID=61935039
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711120851.XA Active CN107948752B (en) | 2017-11-14 | 2017-11-14 | Ordering method, device and terminal for subscription anchor |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107948752B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109086995A (en) * | 2018-08-01 | 2018-12-25 | 广州虎牙信息科技有限公司 | Evaluation method, device, equipment and the storage medium of application program download path |
CN109168049A (en) * | 2018-09-03 | 2019-01-08 | 广州虎牙信息科技有限公司 | Grade evaluation method, device, storage medium and the server of programme televised live |
CN109168048A (en) * | 2018-08-01 | 2019-01-08 | 广州虎牙信息科技有限公司 | Determination method, apparatus, equipment and the storage medium of main broadcaster's grade |
CN109257612A (en) * | 2018-08-09 | 2019-01-22 | 广州虎牙信息科技有限公司 | Assessment Method on Potential, device, computer storage medium and server is broadcast live in game |
CN109409983A (en) * | 2018-09-19 | 2019-03-01 | 北京创鑫旅程网络技术有限公司 | A kind of method and device of ticket type sequence |
CN112312173A (en) * | 2019-07-24 | 2021-02-02 | 广州虎牙科技有限公司 | Anchor recommendation method and device, electronic equipment and readable storage medium |
CN112714329A (en) * | 2020-12-23 | 2021-04-27 | 广州博冠信息科技有限公司 | Display control method and device for live broadcast room, storage medium and electronic equipment |
RU2798452C1 (en) * | 2019-12-18 | 2023-06-22 | Биго Текнолоджи Пте. Лтд. | Method for promotion of information about hosts, computing device and data carrier |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106131584A (en) * | 2016-07-25 | 2016-11-16 | 北京金山安全软件有限公司 | Live broadcast information display method and device and electronic equipment |
CN106604044A (en) * | 2016-11-21 | 2017-04-26 | 武汉斗鱼网络科技有限公司 | Direct broadcast room recommendation strategy assessment method, user equipment, and server system |
CN106878772A (en) * | 2017-02-27 | 2017-06-20 | Ut斯达康(深圳)技术有限公司 | A kind of program commending method and device |
CN106982381A (en) * | 2017-03-31 | 2017-07-25 | 武汉斗鱼网络科技有限公司 | Homepage recommendation process method and device |
CN107172452A (en) * | 2017-04-25 | 2017-09-15 | 北京潘达互娱科技有限公司 | Direct broadcasting room recommends method and device |
CN107205178A (en) * | 2017-04-25 | 2017-09-26 | 北京潘达互娱科技有限公司 | Direct broadcasting room recommends method and device |
-
2017
- 2017-11-14 CN CN201711120851.XA patent/CN107948752B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106131584A (en) * | 2016-07-25 | 2016-11-16 | 北京金山安全软件有限公司 | Live broadcast information display method and device and electronic equipment |
CN106604044A (en) * | 2016-11-21 | 2017-04-26 | 武汉斗鱼网络科技有限公司 | Direct broadcast room recommendation strategy assessment method, user equipment, and server system |
CN106878772A (en) * | 2017-02-27 | 2017-06-20 | Ut斯达康(深圳)技术有限公司 | A kind of program commending method and device |
CN106982381A (en) * | 2017-03-31 | 2017-07-25 | 武汉斗鱼网络科技有限公司 | Homepage recommendation process method and device |
CN107172452A (en) * | 2017-04-25 | 2017-09-15 | 北京潘达互娱科技有限公司 | Direct broadcasting room recommends method and device |
CN107205178A (en) * | 2017-04-25 | 2017-09-26 | 北京潘达互娱科技有限公司 | Direct broadcasting room recommends method and device |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109086995A (en) * | 2018-08-01 | 2018-12-25 | 广州虎牙信息科技有限公司 | Evaluation method, device, equipment and the storage medium of application program download path |
CN109168048A (en) * | 2018-08-01 | 2019-01-08 | 广州虎牙信息科技有限公司 | Determination method, apparatus, equipment and the storage medium of main broadcaster's grade |
CN109168048B (en) * | 2018-08-01 | 2020-12-15 | 广州虎牙信息科技有限公司 | Determination method, device and equipment of anchor grade and storage medium |
CN109257612A (en) * | 2018-08-09 | 2019-01-22 | 广州虎牙信息科技有限公司 | Assessment Method on Potential, device, computer storage medium and server is broadcast live in game |
CN109257612B (en) * | 2018-08-09 | 2020-11-20 | 广州虎牙信息科技有限公司 | Game live broadcast potential evaluation method and device, computer storage medium and server |
CN109168049A (en) * | 2018-09-03 | 2019-01-08 | 广州虎牙信息科技有限公司 | Grade evaluation method, device, storage medium and the server of programme televised live |
CN109409983A (en) * | 2018-09-19 | 2019-03-01 | 北京创鑫旅程网络技术有限公司 | A kind of method and device of ticket type sequence |
CN112312173A (en) * | 2019-07-24 | 2021-02-02 | 广州虎牙科技有限公司 | Anchor recommendation method and device, electronic equipment and readable storage medium |
CN112312173B (en) * | 2019-07-24 | 2023-03-28 | 广州虎牙科技有限公司 | Anchor recommendation method and device, electronic equipment and readable storage medium |
RU2798452C1 (en) * | 2019-12-18 | 2023-06-22 | Биго Текнолоджи Пте. Лтд. | Method for promotion of information about hosts, computing device and data carrier |
CN112714329A (en) * | 2020-12-23 | 2021-04-27 | 广州博冠信息科技有限公司 | Display control method and device for live broadcast room, storage medium and electronic equipment |
CN112714329B (en) * | 2020-12-23 | 2023-09-26 | 广州博冠信息科技有限公司 | Display control method and device for live broadcasting room, storage medium and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN107948752B (en) | 2021-01-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107948752A (en) | Subscribe to main broadcaster's sort method, device and terminal | |
US10405016B2 (en) | Recommending media items based on take rate signals | |
CN105493057B (en) | The content selection controlled using accuracy | |
CN110149540B (en) | Recommendation processing method and device for multimedia resources, terminal and readable medium | |
CN104160711B (en) | User for producing media viewing achievement watches Data Collection | |
CN104090912B (en) | Information-pushing method and device | |
CN107832437B (en) | Audio/video pushing method, device, equipment and storage medium | |
CN108322829A (en) | Personalized main broadcaster recommends method, apparatus and electronic equipment | |
CN111143697B (en) | Content recommendation method and related device | |
CN107038213A (en) | A kind of method and device of video recommendations | |
CN104537000B (en) | A kind of method and apparatus for pushed information | |
CN106294830A (en) | The recommendation method and device of multimedia resource | |
US20150066593A1 (en) | Determining a precision factor for a content selection parameter value | |
CN108363730B (en) | Content recommendation method, system and terminal equipment | |
JP2013088832A (en) | Information processing device, information processing method, and program | |
CN109274987B (en) | Video set ordering method, server and readable storage medium | |
WO2016183936A1 (en) | Application icon arrangement method and device, mobile terminal and storage medium | |
US20170323230A1 (en) | Evaluating keyword performance | |
CN111626779A (en) | Media information processing method, device and storage medium | |
CN109064293A (en) | Method of Commodity Recommendation, device, computer equipment and storage medium | |
CN110347781B (en) | Article reverse arrangement method, article recommendation method, device, equipment and storage medium | |
CN103092348A (en) | Mobile terminal advertisement playing method based on user behavior | |
CN110910201B (en) | Information recommendation control method and device, computer equipment and storage medium | |
CN106454536A (en) | Method and apparatus for determining information recommendation degree | |
CN111107416A (en) | Bullet screen shielding method and device and electronic equipment |
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 |