CN105898421A - Network film and television program intelligent recommendation method and system - Google Patents
Network film and television program intelligent recommendation method and system Download PDFInfo
- Publication number
- CN105898421A CN105898421A CN201510021323.3A CN201510021323A CN105898421A CN 105898421 A CN105898421 A CN 105898421A CN 201510021323 A CN201510021323 A CN 201510021323A CN 105898421 A CN105898421 A CN 105898421A
- Authority
- CN
- China
- Prior art keywords
- movie
- described member
- recommendation
- video programs
- program
- 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.)
- Pending
Links
Abstract
The invention provides a network film and television program intelligent recommendation method, which is characterized in that the method comprises the steps that: an intelligent recommendation system collects information of a member browsing film and television programs and behavior data of the member watching film and television programs of a film and television system; the intelligent recommendation system performs intelligent analysis according to the information of the member browsing film and television programs and the behavior data of the member watching film and television programs, and obtains a film and television program recommendation snapshot of the member; and the film and television program recommendation snapshot of the member is input into the film and television system for use. The invention further discloses a network film and television program intelligent recommendation system. The network film and television program intelligent recommendation method and the network film and television program intelligent recommendation system can perform centralized analysis to obtain a ranking sequence, which is most likely to be watched by the corresponding member, of a massive film and television program library according to historical behavior characteristics of each member and correlation properties between information of the film and television programs, thereby greatly increasing film and television selecting and watching efficiency of the member, and being capable of tapping long-tail economic benefit of the massive film and television program library more fully.
Description
Technical field
Patent of the present invention relates to network and audio frequency and video technical field, particularly relates to one and is applicable to magnanimity movie and video programs, possesses meeting
The online movie program intelligent recommendation method and system of the online movie system of member's characteristic.
Background technology
Along with network technology, video digitization technology and the development of hard-disk storage technology, all kinds of online movie systems is the most emerging
Rise.According to the difference of user's viewing screen, existing online movie system can be divided three classes: the Internet based on computer screen video display
System, mobile shadow viewing system based on mobile phone or flat board and audiovisual shop video system based on the viewing of box projection screen.
The video system of each main flow current is substantially all the movie and video programs storehouse having magnanimity.In traditional cinema market,
The film quantity performed due to the same time is considerably less, and user always can quickly browse all programs and select the film liked
Buy viewing.But the user on network is the most different, owing to being magnanimity library of programmes, the user of the overwhelming majority is substantial amounts of all without spending
Energy travels through all movie and video programs on video system, and the movie and video programs of the most each video system are recommended and search function is the most aobvious
Obtain the most important.
In general, the user of online movie system can be divided into two classes: purpose type user and roaming type user.The former has clearly
Viewing demand it is known that to watch which program, generally use the function of search of video system to find movie and video programs to be watched；
The latter is generally only and at will strolls, and only runs into program interested and is only possible to watch.For current data, the latter to account for
According to bigger proportion.A part the biggest in the latter is member user, and this batch (charge or free) member user repeatedly accesses
After video system, original clear and definite viewing demand is substantially met and gradually transfers roaming type user to.So, how to carry
The viewing rate that browses of high roaming type user is the major issue that each video system is required for considering.
For the Internet video system, owing to the display screen of computer is relatively big, thus would generally use multiple recommendation method (as
Homepage large-size screen monitors, all kinds of ranking list, COLLECTIDN, special topic etc.) and more systematic searching (area, type, time etc.) hands
Section meets the Search Requirement of roaming type user.But this stereotyped and do not distinguish according to user, the most complete to all users
Consistent display content, only has bigger effect to the user of maiden visit, and for those member users (can be the most all
The loyal user repeatedly accessed) for, it is recommended that the program arranging elder generation on position or classification page has been much to watch or be not desired to
Viewing, but the most all come notable position, both cannot improve viewing rate, waste the attention of member.
For mobile shadow viewing system, owing to mobile phone screen and flat screens are narrower, it is recommended that the content of display just with greater need for
Accurately, the user of otherwise roaming type is easy for losing the interest continuing to browse.
Audiovisual shop is a kind of brand-new video display product, and it combines internet mass video display and traditional film institute audiovisual experience two side
Face advantage.User can be to the audiovisual experience of enjoyment movie theatre level in box after chip select, paying.Owing to film each time is watched
Bring the economic benefit of reality will to audiovisual shop, therefore the viewing rate that browses improving user seems outstanding for the video system of audiovisual shop
For important.
On the other hand, video system is owing to having the movie and video programs of magnanimity, and some program may be because of many reasons, always
After being arranged in, this is easily caused long-time the most unmanned it can be found that it, not to mention selects viewing further.How can more have
The rotation of profit shows non-popular program, fully excavates long-tail economic benefit, is also the major issue of all video systems needs consideration.
Summary of the invention
It is desirable to provide a kind of intelligent recommendation method that can be used for each video system, it is possible to according to the history row of each member
It is characterized and associate feature between movie and video programs self-information is right to concentrate analysis to obtain most probable in magnanimity movie and video programs storehouse
Answer what member watched to put in order, be possible not only to substantially increase the chip select viewing efficiency of member, also can excavate magnanimity more fully
The long-tail economic benefit of valut.To achieve these goals, the present invention provides following technical scheme:
A kind of online movie program intelligent recommendation method, described method includes:
(1) on intelligent recommendation systematic collection video system, member browses the information of movie and video programs and member watches the row of movie and video programs
For data；
(2) intelligent recommendation system browses the information of movie and video programs according to described member and described member watches the behavior of movie and video programs
Data carry out intellectual analysis, and the movie and video programs obtaining described member recommend snapshot；
(3) described member recommend that snapshot input video system use.
Preferably, in above-mentioned online movie program intelligent recommendation method, described member is when time viewing movie and video programs behavioral data
The step analyzed includes:
(1) each video system sends described member when time viewing movie and video programs behavior number by the interface of regulation to intelligent recommendation system
According to；
(2) described member is achieved by intelligent recommendation system when time viewing movie and video programs behavioral data；
(3) intelligent recommendation system obtains the historical behavior data of described member, and described member is when time watched movie and video programs phase
Other programme informations closed；
(4) member described in intelligent recommendation system update is when the recommendation exponential model key element value of time watched movie and video programs, then updates
Described member works as the recommendation exponential model key element value of other relevant programs of time watched movie and video programs, then updates other users couple
Described member is when the recommendation exponential model key element value of time watched movie and video programs；
(5) intelligent recommendation system is after completing the renewal of the recommendation exponential model key element value of each program of described member and other users,
The recommendation index that the index Quantitative Calculation Method according to each key element recalculates each program of described member and other users can be further continued for
Value, then according to recommending exponential quantity sequence to obtain the recommendation snapshot that the next access session of described member and other users is to be used.
Preferably, in above-mentioned online movie program intelligent recommendation method, described member is single to be browsed or batch navigation patterns number
Include according to the step analyzed:
(1) each video system sends described member when time viewing movie and video programs behavior number by the interface of regulation to intelligent recommendation system
According to；
(2) described member is achieved by intelligent recommendation system when time viewing movie and video programs behavioral data；
(3) intelligent recommendation system obtains the historical behavior data of described member, and described member is when time watched movie and video programs phase
Other programme informations closed；
(4) member described in intelligent recommendation system update is when the recommendation exponential model key element value of time watched movie and video programs, then updates
Described member works as the recommendation exponential model key element value of other relevant programs of time watched movie and video programs；
(5) intelligent recommendation system is after the recommendation exponential model key element value completing each program of described member updates, and can be further continued for basis
The index Quantitative Calculation Method of each key element recalculates the recommendation exponential quantity of each program of described member, then according to recommending exponential quantity
Sequence obtains the recommendation snapshot that the next access session of described member is to be used.
Preferably, in above-mentioned online movie program intelligent recommendation method, described each video system is quoted in intelligent recommendation system
Recommend snapshot step include:
(1) video system sends, by the interface of regulation, the recommendation snapshot asking to obtain described member to intelligent recommendation system；
(2) intelligent recommendation system checks the conversational list of described member and other users, finds that the nearest session of described member is in effect duration
Within, the most first update the session information of described member, then recommend snapshot storehouse is read this of described member from described member fast
According to, and return SNAPSHOT INFO according to the interface of regulation to each video system；
(3) if the nearest session that intelligent recommendation system inquires described member does not exists or the most expired, then institute can first be write
Stating the up-to-date session information of member, the next snapshot then replicating described member replaces this snapshot of described member, then reads
This snapshot of described member, and return SNAPSHOT INFO according to the interface of regulation to each video system.
Preferably, in above-mentioned online movie program intelligent recommendation method, described index Quantitative Calculation Method passes through at least 2
Index quantitative formula is calculated weighting or the exponential quantity of fall power, and the quantification index value superposition of all key elements obtains pushing away of this program
Recommend index.
Preferably, in above-mentioned online movie program intelligent recommendation method, described program recommends exponential model at least to include following
Key element: ranking, institute in the program that whether described member watched described program, described program was watched in other users 7 days
State program in the program that described member and other users watched ranking, described program by described member the most browsed time
Containing described joint in 10 programs that program several, described has been watched recently by the most browsed number of times of described member, described member
Performer in mesh the classification number of programs of information, described program has performed the joint in 10 programs that described member watches recently simultaneously
Director in mesh quantity, described program has directed the number of programs in 10 programs that described member watches recently, described joint simultaneously
Produce serial number in generic series sheet of time, described program, described member of purpose has seen the series in generic series sheet recently
Number.
Preferably, in above-mentioned online movie program intelligent recommendation method, described member is the recommendation of each movie and video programs in recommending snapshot
The calculation criterion of index includes:
(1) the recommendation index to the movie and video programs that described member had watched makees the process of fall power；
(2) the recommendation index of the popular movie and video programs watched other users in the recent period makees weighting process；
(3) the recommendation index to the popular movie and video programs on the general list list of described member and other users makees weighting process；
(4) the recommendation index of other movie and video programs in classification belonging to the movie and video programs having already viewed by described member is made at weighting
Reason；
(5) make to add to the recommendation index of the performer of the movie and video programs that described member has already viewed by, other movie and video programs that director is relevant
Power processes；
(6) recommend index to make the most weightings to process to time the nearest movie and video programs；
(7) if the movie and video programs that described member watched belong to one of certain series performance, then the recommendation to its follow-up series performance
Index is made weighting and is processed；The recommendation index of the series performance that continues before it is made weighting process；
(8) the recommendation index of the movie and video programs the most browsed to described member makees the process of fall power；
(9) the recommendation index of the movie and video programs the most browsed to described member makees the process of fall power.
Additionally, present invention also offers a kind of online movie program intelligent recommendation system, described movie and video programs commending system includes:
Video system, for providing movie and video programs and member the information browsing movie and video programs, and can carry to member according to recommendation results
For program；
Collection module, browses the information of movie and video programs and member watches the behavior of movie and video programs for collecting member on video system
Data；
Intelligent analysis module, the information and described member for browsing movie and video programs according to described member watches the row of movie and video programs
Intellectual analysis is carried out for data；
Recommend snapshot storehouse, recommend snapshot for generating, update and preserve the movie and video programs of described member.
Online movie program intelligent recommendation method and system provided by the present invention, it is possible to the historical behavior according to each member is special
Levy and associate feature between movie and video programs self-information is to concentrate analysis to obtain in magnanimity movie and video programs storehouse most probable by correspondence meeting
Putting in order of member's viewing.Therefore the present invention has the following technical effect that
(1) can provide according to " the recommendation snapshot " obtained after its behavior hobby intellectual analysis for each member, be substantially improved
Member retrieves experience and raising member browses viewing rate；
(2) raising browsing viewing rate can bring the economic benefit of reality equally to video system；
(3) good non-big movie rotation proposed algorithm can allow each movie and video programs have the opportunity to all members " to see " and arrive,
Fully excavate the economic benefit of each movie and video programs, improve overall long-tail economic benefit.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the online movie program intelligent recommendation method of the present invention；
Fig. 2 is that the movie and video programs of the member of the embodiment of the present invention recommend exponential model structural representation；
Fig. 3 is the schematic flow sheet of the behavioral data analysis of the present invention；
Fig. 4 is the schematic flow sheet quoting intelligent recommendation snapshot of the present invention.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely retouched
State, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments.Based on the present invention
In embodiment, the every other embodiment that those of ordinary skill in the art are obtained under not making creative work premise,
Broadly fall into the scope of protection of the invention.
Embodiment 1
As it is shown in figure 1, a kind of online movie program intelligent recommendation method of the present invention, described method includes:
(1) on intelligent recommendation systematic collection video system, member browses the information of movie and video programs and member watches the row of movie and video programs
For data, member of the present invention can use technological means uniquely to identify；
(2) intelligent recommendation system browses the information of movie and video programs according to described member and described member watches the behavior of movie and video programs
Data carry out intellectual analysis, and the movie and video programs obtaining described member recommend snapshot；
(3) described member recommend that snapshot input video system use.
The principle of the online movie program intelligent recommendation method work of the present invention is, it is possible to according to the historical behavior of each member
It is corresponding that associate feature between feature and movie and video programs self-information concentrates analysis to obtain most probable in magnanimity movie and video programs storehouse
Putting in order of member's viewing.The information of described movie and video programs specifically includes the classification of program, performer, director, the time, affiliated
Series etc..Described member watches the behavioral data of movie and video programs and includes that member browses single movie and video programs behavioral data and member is same
Time browse the behavioral data of multiple movie and video programs, but member will not be by newly generated behavior when the snapshot of recommending in time access session
The impact of data, member when time browse, viewing behavior only influences whether the recommendation snapshot in the access session next time of this member.
Access session of described member be defined as starting from member accesses system until " next operation the most at the appointed time (as
2 hours) occur " time terminate.
From technique scheme it can be seen that the online movie program intelligent recommendation method of the present invention, can carry for each member
For according to after its behavior hobby intellectual analysis " the recommendation snapshot " that obtain, it is substantially improved member and retrieves experience and improve member and browse
Viewing rate；The raising browsing viewing rate can bring the economic benefit of reality equally to video system；Meanwhile, good non-big movie
Rotation proposed algorithm can allow each movie and video programs have the opportunity to all members " to see " and arrive, and fully excavates each video display joint
Purpose economic benefit, improves overall long-tail economic benefit.
Embodiment 2
As shown in Figure 1, Figure 3, a kind of online movie program intelligent recommendation method of the present invention, including step:
(1) on intelligent recommendation systematic collection video system, member browses the information of movie and video programs and member watches the row of movie and video programs
For data；
(2) each video system sends described member when time viewing movie and video programs behavior number by the interface of regulation to intelligent recommendation system
According to；
(3) described member is achieved by intelligent recommendation system when time viewing movie and video programs behavioral data；
(4) intelligent recommendation system obtains the historical behavior data of described member, and described member is when time watched movie and video programs phase
Other programme informations closed；
(5) member described in intelligent recommendation system update is when the recommendation exponential model key element value of time watched movie and video programs, then updates
Described member works as the recommendation exponential model key element value of other relevant programs of time watched movie and video programs, then updates other users couple
Described member is when the recommendation exponential model key element value of time watched movie and video programs；
(6) intelligent recommendation system is after completing the renewal of the recommendation exponential model key element value of each program of described member and other users,
The recommendation index that the index Quantitative Calculation Method according to each key element recalculates each program of described member and other users can be further continued for
Value, then according to recommending exponential quantity sequence to obtain the recommendation snapshot that the next access session of described member and other users is to be used；
(7) described member recommend that snapshot input video system use.
Embodiment 3
As shown in Figure 1, Figure 3, a kind of online movie program intelligent recommendation method of the present invention, including step:
(1) on intelligent recommendation systematic collection video system, member browses the information of movie and video programs and member watches the row of movie and video programs
For data；
(2) each video system sends described member when time viewing movie and video programs behavior number by the interface of regulation to intelligent recommendation system
According to；
(3) described member is achieved by intelligent recommendation system when time viewing movie and video programs behavioral data；
(4) intelligent recommendation system obtains the historical behavior data of described member, and described member is when time watched movie and video programs phase
Other programme informations closed；
(5) member described in intelligent recommendation system update is when the recommendation exponential model key element value of time watched movie and video programs, then updates
Described member works as the recommendation exponential model key element value of other relevant programs of time watched movie and video programs；
(6) intelligent recommendation system is after the recommendation exponential model key element value completing each program of described member updates, and can be further continued for basis
The index Quantitative Calculation Method of each key element recalculates the recommendation exponential quantity of each program of described member, then according to recommending exponential quantity
Sequence obtains the recommendation snapshot that the next access session of described member is to be used；
(7) described member recommend that snapshot input video system use.
Embodiment 4
As Figure 1 and Figure 4, a kind of online movie program intelligent recommendation method of the present invention, including step:
(1) on intelligent recommendation systematic collection video system, member browses the information of movie and video programs and member watches the row of movie and video programs
For data；
(2) intelligent recommendation system browses the information of movie and video programs according to described member and described member watches the behavior of movie and video programs
Data carry out intellectual analysis, and the movie and video programs obtaining described member recommend snapshot；
(3) video system sends, by the interface of regulation, the recommendation snapshot asking to obtain described member to intelligent recommendation system；
(4) intelligent recommendation system checks the conversational list of described member and other users, finds that the nearest session of described member is in effect duration
Within, the most first update the session information of described member, then recommend snapshot storehouse is read this of described member from described member fast
According to, and return SNAPSHOT INFO according to the interface of regulation to each video system；
(5) if the nearest session that intelligent recommendation system inquires described member does not exists or the most expired, then institute can first be write
Stating the up-to-date session information of member, the next snapshot then replicating described member replaces this snapshot of described member, then reads
This snapshot of described member, and return SNAPSHOT INFO according to the interface of regulation to each video system；
(6) described member recommend that snapshot input video system use.
Embodiment 5
As Figure 1 and Figure 4, a kind of online movie program intelligent recommendation system approach of the present invention, described index quantum chemical method
Method is calculated weighting or the exponential quantity of fall power by least 2 index quantitative formulas, and the quantification index value of all key elements is folded
Add the recommendation index obtaining this program.
Embodiment 6
As shown in Figure 1 and Figure 2, a kind of online movie program intelligent recommendation system approach of the present invention, including step:
(1) on intelligent recommendation systematic collection video system, member browses the information of movie and video programs and member watches the row of movie and video programs
For data；
(2) each video system sends described member when time viewing movie and video programs behavior number by the interface of regulation to intelligent recommendation system
According to；
(3) described member is achieved by intelligent recommendation system when time viewing movie and video programs behavioral data；
(4) intelligent recommendation system obtains the historical behavior data of described member, and described member is when time watched movie and video programs phase
Other programme informations closed；
(5) member described in intelligent recommendation system update is when the recommendation exponential model key element value of time watched movie and video programs, then updates
Described member works as the recommendation exponential model key element value of other relevant programs of time watched movie and video programs, then updates other users couple
Described member is when the recommendation exponential model key element value of time watched movie and video programs；
(6) intelligent recommendation system is after completing the renewal of the recommendation exponential model key element value of each program of described member and other users,
The recommendation index that the index Quantitative Calculation Method according to each key element recalculates each program of described member and other users can be further continued for
Value, then according to recommending exponential quantity sequence to obtain the recommendation snapshot that the next access session of described member and other users is to be used；
Described program recommends exponential model at least to include following key element: whether described member watched described program, described program exists
In the program that other users watched in 7 days ranking, described program in the program that described member and other users watched ranking,
Described program by the most browsed number of times of described member, described program by the most browsed number of times of described member, described
10 programs that member watches recently are performed containing the performer in the number of programs of described programme labeling information, described program simultaneously
Number of programs in 10 programs that described member watches recently, that the director in described program has directed described member simultaneously is nearest
Number of programs in 10 programs of viewing, the serial number in generic series sheet of time, described program of producing of described program,
Described member has seen the serial number in generic series sheet recently；
(7) described member recommend that snapshot input video system use.
Embodiment 7
As Figure 1 and Figure 4, a kind of online movie program intelligent recommendation system approach of the present invention, including step:
(1) on intelligent recommendation systematic collection video system, member browses the information of movie and video programs and member watches the row of movie and video programs
For data；
(2) each video system sends described member when time viewing movie and video programs behavior number by the interface of regulation to intelligent recommendation system
According to；
(3) described member is achieved by intelligent recommendation system when time viewing movie and video programs behavioral data；
(4) intelligent recommendation system obtains the historical behavior data of described member, and described member is when time watched movie and video programs phase
Other programme informations closed；
(5) member described in intelligent recommendation system update is when the recommendation exponential model key element value of time watched movie and video programs, then updates
Described member works as the recommendation exponential model key element value of other relevant programs of time watched movie and video programs, then updates other users couple
Described member is when the recommendation exponential model key element value of time watched movie and video programs；
(6) intelligent recommendation system is after completing the renewal of the recommendation exponential model key element value of each program of described member and other users,
The recommendation index that the index Quantitative Calculation Method according to each key element recalculates each program of described member and other users can be further continued for
Value, then according to recommending exponential quantity sequence to obtain the recommendation snapshot that the next access session of described member and other users is to be used；
Described member in recommending snapshot each movie and video programs recommend index calculation criterion include: the shadow that described member had been watched
Make fall power depending on the recommendation index of program to process；The recommendation index of the popular movie and video programs watched other users in the recent period is made at weighting
Reason；The recommendation index of the popular movie and video programs on the general list list of described member and other users is made weighting process；To described meeting
The recommendation index of other movie and video programs in classification belonging to the movie and video programs that member has already viewed by is made weighting and is processed；Described member is seen
The recommendation index of other movie and video programs that the performer of the movie and video programs seen, director are correlated with is made weighting and is processed；The shadow the nearest to the time
The weighting the most depending on the recommendation index work of program processes；If the movie and video programs that described member watched belong to one of certain series performance,
Then the recommendation index to its follow-up series performance makees weighting process；The recommendation index of the series performance that continues before it is made weighting process；Right
The recommendation index of the movie and video programs that described member is the most browsed is made fall power and is processed；The movie and video programs the most browsed to described member
Recommendation index make fall power process；
(7) described member recommend that snapshot input video system use.
Embodiment 8
As it is shown in figure 1, a kind of online movie program intelligent recommendation system of the present invention, described movie and video programs commending system includes:
Video system, for providing movie and video programs and member the information browsing movie and video programs, and can carry to member according to recommendation results
For program；
Collection module, browses the information of movie and video programs and member watches the behavior of movie and video programs for collecting member on video system
Data；
Intelligent analysis module, the information and described member for browsing movie and video programs according to described member watches the row of movie and video programs
Intellectual analysis is carried out for data；
Recommend snapshot storehouse, recommend snapshot for generating, update and preserve the movie and video programs of described member.
This system can collect according to the associate feature between the historical behavior feature of each member and movie and video programs self-information
Middle analysis obtains most probable putting in order by corresponding member's viewing in magnanimity movie and video programs storehouse, is possible not only to substantially increase member
Chip select viewing efficiency, also can excavate the long-tail economic benefit of magnanimity valut more fully.
Claims (8)
1. an online movie program intelligent recommendation method, it is characterised in that described method includes:
(1) on intelligent recommendation systematic collection video system, member browses the information of movie and video programs and member watches the row of movie and video programs
For data；
(2) intelligent recommendation system browses the information of movie and video programs according to described member and described member watches the behavior of movie and video programs
Data carry out intellectual analysis, and the movie and video programs obtaining described member recommend snapshot；
(3) described member recommend that snapshot input video system use.
Online movie program intelligent recommendation method the most according to claim 1, it is characterised in that described member is when time viewing video display
The step of program behavioral data analysis includes:
(1) each video system sends described member when time viewing movie and video programs behavior number by the interface of regulation to intelligent recommendation system
According to；
(2) described member is achieved by intelligent recommendation system when time viewing movie and video programs behavioral data；
(3) intelligent recommendation system obtains the historical behavior data of described member, and described member is when time watched movie and video programs phase
Other programme informations closed；
(4) member described in intelligent recommendation system update is when the recommendation exponential model key element value of time watched movie and video programs, then updates
Described member works as the recommendation exponential model key element value of other relevant programs of time watched movie and video programs, then updates other users couple
Described member is when the recommendation exponential model key element value of time watched movie and video programs；
(5) intelligent recommendation system is after completing the renewal of the recommendation exponential model key element value of each program of described member and other users,
The recommendation index that the index Quantitative Calculation Method according to each key element recalculates each program of described member and other users can be further continued for
Value, then according to recommending exponential quantity sequence to obtain the recommendation snapshot that the next access session of described member and other users is to be used.
Online movie program intelligent recommendation method the most according to claim 1, it is characterised in that described member is single to be browsed or criticize
The step of amount navigation patterns data analysis includes:
(1) each video system sends described member when time viewing movie and video programs behavior number by the interface of regulation to intelligent recommendation system
According to；
(2) described member is achieved by intelligent recommendation system when time viewing movie and video programs behavioral data；
(3) intelligent recommendation system obtains the historical behavior data of described member, and described member is when time watched movie and video programs phase
Other programme informations closed；
(4) member described in intelligent recommendation system update is when the recommendation exponential model key element value of time watched movie and video programs, then updates
Described member works as the recommendation exponential model key element value of other relevant programs of time watched movie and video programs；
(5) intelligent recommendation system is after the recommendation exponential model key element value completing each program of described member updates, and can be further continued for basis
The index Quantitative Calculation Method of each key element recalculates the recommendation exponential quantity of each program of described member, then according to recommending exponential quantity
Sequence obtains the recommendation snapshot that the next access session of described member is to be used.
Online movie program intelligent recommendation method the most according to claim 1, it is characterised in that described each video system quotes intelligence
The step recommending snapshot on energy commending system includes:
(1) video system sends, by the interface of regulation, the recommendation snapshot asking to obtain described member to intelligent recommendation system；
(2) intelligent recommendation system checks the conversational list of described member and other users, finds that the nearest session of described member is in effect duration
Within, the most first update the session information of described member, then recommend snapshot storehouse is read this of described member from described member fast
According to, and return SNAPSHOT INFO according to the interface of regulation to each video system；
(3) if the nearest session that intelligent recommendation system inquires described member does not exists or the most expired, then institute can first be write
Stating the up-to-date session information of member, the next snapshot then replicating described member replaces this snapshot of described member, then reads
This snapshot of described member, and return SNAPSHOT INFO according to the interface of regulation to each video system.
Online movie program intelligent recommendation method the most according to claim 2, it is characterised in that described index Quantitative Calculation Method
Being calculated weighting or the exponential quantity of fall power by least 2 index quantitative formulas, the quantification index value superposition of all key elements obtains
Recommendation index to this program.
Online movie program intelligent recommendation method the most according to claim 2, it is characterised in that described program recommends exponential model
At least include following key element: the joint that whether described member watched described program, described program was watched in other users 7 days
In mesh, ranking, described program ranking, described program in the program that described member and other users watched are thick by described member
10 programs that the most browsed number of times, described program have been watched recently by the most browsed number of times of described member, described member
In performed that described member watches recently containing the performer in the number of programs of described programme labeling information, described program simultaneously 10
Number of programs in portion's program, the director in described program have directed the program in 10 programs that described member watches recently simultaneously
Quantity, time, the described program serial number in generic series sheet of producing of described program, described member have seen affiliated system recently
Serial number in column-slice.
Online movie program intelligent recommendation method the most according to claim 2, it is characterised in that described member is each in recommending snapshot
Movie and video programs recommend the calculation criterion of index to include:
(1) the recommendation index to the movie and video programs that described member had watched makees the process of fall power；
(2) the recommendation index of the popular movie and video programs watched other users in the recent period makees weighting process；
(3) the recommendation index to the popular movie and video programs on the general list list of described member and other users makees weighting process；
(4) the recommendation index of other movie and video programs in classification belonging to the movie and video programs having already viewed by described member is made at weighting
Reason；
(5) make to add to the recommendation index of the performer of the movie and video programs that described member has already viewed by, other movie and video programs that director is relevant
Power processes；
(6) recommend index to make the most weightings to process to time the nearest movie and video programs；
(7) if the movie and video programs that described member watched belong to one of certain series performance, then the recommendation to its follow-up series performance
Index is made weighting and is processed；The recommendation index of the series performance that continues before it is made weighting process；
(8) the recommendation index of the movie and video programs the most browsed to described member makees the process of fall power；
(9) the recommendation index of the movie and video programs the most browsed to described member makees the process of fall power.
8. an online movie program intelligent recommendation system, it is characterised in that described movie and video programs commending system includes:
Video system, for providing movie and video programs and member the information browsing movie and video programs, and can carry to member according to recommendation results
For program；
Collection module, browses the information of movie and video programs and member watches the behavior of movie and video programs for collecting member on video system
Data；
Intelligent analysis module, the information and described member for browsing movie and video programs according to described member watches the row of movie and video programs
Intellectual analysis is carried out for data；
Recommend snapshot storehouse, recommend snapshot for generating, update and preserve the movie and video programs of described member.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510021323.3A CN105898421A (en) | 2015-01-16 | 2015-01-16 | Network film and television program intelligent recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510021323.3A CN105898421A (en) | 2015-01-16 | 2015-01-16 | Network film and television program intelligent recommendation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105898421A true CN105898421A (en) | 2016-08-24 |
Family
ID=56999044
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510021323.3A Pending CN105898421A (en) | 2015-01-16 | 2015-01-16 | Network film and television program intelligent recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105898421A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107180374A (en) * | 2017-04-28 | 2017-09-19 | 百度在线网络技术（北京）有限公司 | A kind of method and apparatus that member is customized for user |
| CN108363730A (en) * | 2018-01-12 | 2018-08-03 | 优地网络有限公司 | A kind of content recommendation method, system and terminal device |
| CN109922369A (en) * | 2017-12-12 | 2019-06-21 | 深圳Tcl新技术有限公司 | A kind of smart television program commending method, system and storage medium |
| CN113139856A (en) * | 2021-05-14 | 2021-07-20 | 青岛聚看云科技有限公司 | Movie and television member package recommendation method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101141607A (en) * | 2006-09-08 | 2008-03-12 | 百视通网络电视技术发展有限责任公司 | Mutual correlation method for IPTV and implementing system thereof |
| CN101764661A (en) * | 2008-12-23 | 2010-06-30 | 未序网络科技（上海）有限公司 | Data fusion based video program recommendation system |
| CN102256169A (en) * | 2010-05-21 | 2011-11-23 | 腾讯科技（深圳）有限公司 | Method and device for recommending relevant videos to users |
| CN102685566A (en) * | 2012-05-28 | 2012-09-19 | 北京网尚数字电影院线有限公司 | Recommendation method for audio and video programs |
-
2015
- 2015-01-16 CN CN201510021323.3A patent/CN105898421A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101141607A (en) * | 2006-09-08 | 2008-03-12 | 百视通网络电视技术发展有限责任公司 | Mutual correlation method for IPTV and implementing system thereof |
| CN101764661A (en) * | 2008-12-23 | 2010-06-30 | 未序网络科技（上海）有限公司 | Data fusion based video program recommendation system |
| CN102256169A (en) * | 2010-05-21 | 2011-11-23 | 腾讯科技（深圳）有限公司 | Method and device for recommending relevant videos to users |
| CN102685566A (en) * | 2012-05-28 | 2012-09-19 | 北京网尚数字电影院线有限公司 | Recommendation method for audio and video programs |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107180374A (en) * | 2017-04-28 | 2017-09-19 | 百度在线网络技术（北京）有限公司 | A kind of method and apparatus that member is customized for user |
| CN109922369A (en) * | 2017-12-12 | 2019-06-21 | 深圳Tcl新技术有限公司 | A kind of smart television program commending method, system and storage medium |
| CN108363730A (en) * | 2018-01-12 | 2018-08-03 | 优地网络有限公司 | A kind of content recommendation method, system and terminal device |
| CN108363730B (en) * | 2018-01-12 | 2020-11-06 | 优地网络有限公司 | Content recommendation method, system and terminal equipment |
| CN113139856A (en) * | 2021-05-14 | 2021-07-20 | 青岛聚看云科技有限公司 | Movie and television member package recommendation method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2817970B1 (en) | Automatically recommending content | |
| CN105404700B (en) | A kind of video column recommendation system and recommended method based on collaborative filtering | |
| US20190279274A1 (en) | Intelligent offer for related products with preview and real time feedback | |
| CN102685566A (en) | Recommendation method for audio and video programs | |
| CN107980129B (en) | Global recommendation system for overlapping media directories | |
| CN103747343B (en) | The method and apparatus that resource is recommended at times | |
| US9681168B2 (en) | Recommending a composite channel | |
| KR101832996B1 (en) | Enhanced program guide | |
| US9235574B2 (en) | Systems and methods for providing media recommendations | |
| US9674576B2 (en) | Methods and systems of providing a supplemental experience based on concurrently viewed content | |
| CN105359125B (en) | User's history playlist and reservation | |
| EP2835979A1 (en) | Attribute setting method and device | |
| CN105915949A (en) | Video content recommending method, device and system | |
| US20120179717A1 (en) | System and method for effectively providing entertainment recommendations to device users | |
| CN105095431A (en) | Method and device for pushing videos based on behavior information of user | |
| CN104065981A (en) | Method and device for recommending videos | |
| CN101271559A (en) | Cooperation recommending system based on user partial interest digging | |
| CN111866528A (en) | Live program pushing method and readable storage medium | |
| CN103686231A (en) | Method and system for integrated management, failure replacement and continuous playing of film | |
| JP2008542870A (en) | Method and apparatus for estimating the overall interest of a group of users for content | |
| CN103634687A (en) | Method and system of providing video retrieval results in intelligent television | |
| CN105898421A (en) | Network film and television program intelligent recommendation method and system | |
| CN104866591A (en) | Video recommendation method and device | |
| CN103369374B (en) | Method and apparatus for operation content channel | |
| WO2014051853A1 (en) | Optimizing monetization with brand impact scoring |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160824 |