CN109657138A - A kind of video recommendation method, device, electronic equipment and storage medium - Google Patents
A kind of video recommendation method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN109657138A CN109657138A CN201811507325.3A CN201811507325A CN109657138A CN 109657138 A CN109657138 A CN 109657138A CN 201811507325 A CN201811507325 A CN 201811507325A CN 109657138 A CN109657138 A CN 109657138A
- Authority
- CN
- China
- Prior art keywords
- video
- behavior
- video tab
- user
- tab
- 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
Abstract
The present invention relates to a kind of video recommendation method, device, electronic equipment and storage mediums, belong to internet art field.The video recommendation method includes: to determine multiple video tabs according to the consumption log of user, wherein the video tab is used to indicate the classification of video;The user is calculated to the preference value of each video tab in the multiple video tab;It is determined according to the preference value of each video tab and recommends video, and recommended to client.In the embodiment of the present application, pass through the consumption log of user, determine the corresponding video tab of multiple videos of the customer consumption, the user is calculated again to the preference value of each video tab, then it is determined according to calculated result and recommends video, namely the video recommended according to the preference of user to it, it is ensured that its interested video can be recommended for user.
Description
Technical field
The invention belongs to internet art fields, and in particular to a kind of video recommendation method, device, electronic equipment and storage are situated between
Matter.
Background technique
With the development of internet technology, the channel of people's acquisition information is more and more, and all kinds of platforms provide for user
Including information such as article, atlas, videos, for platform, how to push the interested information of user to user is platform head
The problem of facing.Platform information recommended to the user is often determined by all kinds of algorithms, and algorithm is needed by mass data
Training and test can just accurately determine the information that different user is liked, since the required data volume of, the training of algorithm is huge
Greatly, often daylong data are trained with one day time interval, tested, therefore, the message recommendation carried out with this does not have
It the operation behavior current to user can feed back in time, be often required to just calculate the interested recommendation of user by second day
Content, so that recommendation effect is undesirable.
Summary of the invention
In consideration of it, the purpose of the present invention is to provide a kind of video recommendation method, device, electronic equipment and storage medium,
To effectively improve the above problem.
The embodiment of the present invention is achieved in that
In a first aspect, the embodiment of the invention provides a kind of video recommendation methods, comprising: true according to the consumption log of user
Make multiple video tabs, wherein the video tab is used to indicate the classification of video;The user is calculated to the multiple view
The preference value of each video tab in frequency marking label;It is determined according to the preference value of each video tab and recommends video, and pushed away
It recommends to client.
A kind of possible embodiment with reference to first aspect, calculates the user to every in the multiple video tab
The preference value of a video tab, comprising: obtain the last user calculated to each view in the multiple video tab
The history preference value of frequency marking label;The acquisition last time calculates time point to the time interval of current point in time, and according to the time
It is spaced and determines time attenuation coefficient;Calculate all behaviors of the user to each video tab in the multiple video tab
Weight;Determine the user to described according to the history preference value, the time attenuation coefficient and all behavior weights
The preference value of each video tab in multiple video tabs.
A kind of possible embodiment with reference to first aspect, according to the history preference value, the time attenuation coefficient
Determine the user to the preference value of each video tab in the multiple video tab, packet with all behavior weights
It includes: according to the history preference value, the time attenuation coefficient and all behavior weights and preference formula P (Ui, Tj)
=P (Ui, Tj) ' * TimeDecay+SumAllAction (ActionWeight*ItemWeight*ItemTagWeight*
UserTagWeight) determine the user to the preference value of each video tab in the multiple video tab;Wherein, P
(Ui, Tj) indicates user i to the preference value of video tab j;P (Ui, Tj) ' indicates last calculated user i to video mark
The history preference value for signing j, calculating then value if first time is zero;TimeDecay expression time attenuation coefficient=exp (-
DeltaSeconds/decayFactor), time of the deltaSeconds expression last calculating time point to current point in time
Interval, unit is the second;DecayFactor is right for positive preference decayFactor=800000 for controlling rate of decay
In negative sense preference decayFactor=100000;SumAllAction is all behavior weights comprising video tab j, that is, is wrapped
The sum of include click, broadcasting, downloading, expose, do not like five kinds of behavior weights;ActionWeight is single behavior weight, for
Positive preference, single behavior weight are respectively as follows: click behavior=1, play behavior=2, download behavior=5, and exposure behavior=-
0.1, behavior=- 5 are not liked, for negative sense preference, single behavior weight is respectively as follows: click behavior=- 1, plays behavior
Behavior=- 5 are downloaded in=- 2, expose behavior=1, do not like behavior=5;ItemWeight indicates the weight of video tab j,
ItemWeight=log (ActionCount/ItemActionCount), wherein ActionCount indicates that all users are directed to
The sum of some behavior of video tab j, ItemActionCount indicate that user i is total under the behavior for video tab j
Number;ItemTagWeight indicates weight of the Tj in Item, is default value;UserTagWeight indicates Tj in user preference
Weight, (all video tab history tire out calculated number/video tab j cumulative calculation mistake to UserTagWeight=log
Number).
A kind of possible embodiment with reference to first aspect, the consumption log are search log;According to disappearing for user
Multiple video tabs are determined in expense log, comprising: word cutting processing is carried out as unit of word to the search term in described search log,
Obtain multiple words;Duplicate removal processing is carried out to the multiple word;Permutation and combination is carried out to multiple words after duplicate removal, obtains multiple words;
By each word in the multiple word respectively with video library tag match, using with the word of the video library tag hit as described in
Video tab.
A kind of possible embodiment with reference to first aspect determines according to the preference value of each video tab and recommends
Video, comprising: promote decision-tree model using trained gradient and the preference value of each video tab is ranked up, obtain
To ranking results;The recommendation video is determined according to the ranking results.
Second aspect, the embodiment of the present application also provides a kind of video recommendations devices, comprising: the first determining module is used for
Multiple video tabs are determined according to the consumption log of user, wherein the video tab is used to indicate the classification of video;It calculates
Module, for calculating the user to the preference value of each video tab in the multiple video tab;Second determining module,
Recommend video for determining according to the preference value of each video tab, and recommends to client.
In conjunction with a kind of possible embodiment of second aspect, the computing module is also used to obtain last calculating
History preference value of the user to each video tab in the multiple video tab;Acquisition last calculating time point arrives
The time interval of current point in time, and time attenuation coefficient is determined according to the time interval;The user is calculated to described more
All behavior weights of each video tab in a video tab;According to the history preference value, the time attenuation coefficient
Determine the user to the preference value of each video tab in the multiple video tab with all behavior weights.
In conjunction with a kind of possible embodiment of second aspect, the computing module is also used to according to the history preference
Value, the time attenuation coefficient and all behavior weights and preference formula P (Ui, Tj)=P (Ui, Tj) ' *
TimeDecay+SumAllAction (ActionWeight*ItemWeight*ItemTagWeight*UserTagWeight) is true
Preference value of the fixed user to each video tab in the multiple video tab;Wherein, P (Ui, Tj) indicates i pairs of user
The preference value of video tab j;P (Ui, Tj) ' indicate last calculated user i to the history preference value of video tab j, if
Calculating then value for first time is zero;TimeDecay indicates time attenuation coefficient=exp (- deltaSeconds/
DecayFactor), deltaSeconds indicates last and calculates time point to the time interval of current point in time, and unit is the second;
DecayFactor is for controlling rate of decay, for positive preference decayFactor=800000, for negative sense preference
DecayFactor=100000;SumAllAction is all behavior weights comprising video tab j, that is, includes clicking, broadcasting
The sum of put, download, exposing, not liking five kinds of behavior weights;ActionWeight is single behavior weight, for positive preference,
Single behavior weight is respectively as follows: click behavior=1, plays behavior=2, downloads behavior=5, exposes behavior=- 0.1, does not like
Behavior=- 5, for negative sense preference, single behavior weight is respectively as follows: click behavior=- 1, plays behavior=- 2, downloading row
It is=- 5, exposes behavior=1, do not like behavior=5;ItemWeight indicates the weight of video tab j, wherein
ActionCount indicates that the sum of some behavior of all users for video tab j, ItemActionCount indicate user i
For sum of the video tab j under the behavior;ItemTagWeight indicates weight of the Tj in Item, is default value;
UserTagWeight indicates weight of the Tj in user preference, and (all video tab history are accumulative by UserTagWeight=log
The number that the number calculated/video tab j cumulative calculation is crossed).
In conjunction with a kind of possible embodiment of second aspect, the consumption log is to search for log, the first determining module,
It is also used to carry out the search term in described search log as unit of word word cutting processing, obtains multiple words;To the multiple word
Carry out duplicate removal processing;Permutation and combination is carried out to multiple words after duplicate removal, obtains multiple words;By each word in the multiple word point
Not with video library tag match, using with the word of the video library tag hit as the video tab.
In conjunction with a kind of possible embodiment of second aspect, the second determining module is also used to utilize trained gradient
It promotes decision-tree model to be ranked up the preference value of each video tab, obtains ranking results;It is tied according to the sequence
Fruit determines the recommendation video.
The third aspect, the embodiment of the present application also provides a kind of electronic equipment, comprising: memory and processor, it is described to deposit
Reservoir is connected with the processor;The memory, for storing program;The processor is stored in described deposit for calling
Program in reservoir, to execute the side of above-mentioned first aspect and/or a kind of possible embodiment offer with reference to first aspect
Method.
Fourth aspect, the embodiment of the present application also provides a kind of storage medium, including computer program, the computer journey
The side of above-mentioned first aspect and/or a kind of possible embodiment offer with reference to first aspect is provided when sequence is run by processor
Method.
Video recommendation method provided in an embodiment of the present invention, comprising: multiple videos are determined according to the consumption log of user
Label, wherein the video tab is used to indicate the classification of video;The user is calculated to every in the multiple video tab
The preference value of a video tab;It is determined according to the preference value of each video tab and recommends video, and recommended to client.This
Apply by the consumption log of user, determining the corresponding video tab of multiple videos of the customer consumption, then count in embodiment
The user is calculated to the preference value of each video tab, is then determined according to calculated result and recommends video, namely according to the inclined of user
The good video recommended to it, it is ensured that its interested video can be recommended for user.
Other features and advantages of the present invention will be illustrated in subsequent specification, also, partly be become from specification
It is clear that being understood by implementing the embodiment of the present invention.The objectives and other advantages of the invention can be by written
Specifically noted structure is achieved and obtained in specification, claims and attached drawing.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings
Obtain other attached drawings.By the way that shown in attached drawing, above and other purpose of the invention, feature and advantage will be more clear.In whole
Identical appended drawing reference indicates identical part in attached drawing.Attached drawing, emphasis deliberately are not drawn by actual size equal proportion scaling
It is to show the gist of the present invention.
Fig. 1 shows the structural schematic diagram of a kind of electronic equipment provided in an embodiment of the present invention.
Fig. 2 shows a kind of flow diagrams of video recommendation method provided in an embodiment of the present invention.
Fig. 3 shows the flow diagram of the step S102 in Fig. 2 provided in an embodiment of the present invention.
Fig. 4 shows a kind of module diagram of video recommendations device provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.The present invention being usually described and illustrated herein in the accompanying drawings is implemented
The component of example can be arranged and be designed with a variety of different configurations.
Therefore, the detailed description of the embodiment of the present invention provided in the accompanying drawings is not intended to limit below claimed
The scope of the present invention, but be merely representative of selected embodiment of the invention.Based on the embodiments of the present invention, this field is common
Technical staff's every other embodiment obtained without creative efforts belongs to the model that the present invention protects
It encloses.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.
In the description of the present invention, it should be noted that term " first ", " second ", " third " etc. are only used for distinguishing and retouch
It states, is not understood to indicate or imply relative importance.Furthermore term "and/or" in the application, only a kind of description is closed
Join the incidence relation of object, indicates may exist three kinds of relationships, for example, A and/or B, can indicate: individualism A is deposited simultaneously
In A and B, these three situations of individualism B.
As shown in Figure 1, Fig. 1 shows the structural block diagram of a kind of electronic equipment 100 provided in an embodiment of the present invention.The electricity
Sub- equipment 100 includes: video recommendations device 110, memory 120, storage control 130 and processor 140.Wherein, in this hair
In bright embodiment, the electronic equipment 100 may be, but not limited to, network server, database server, cloud server
Deng.
The memory 120, storage control 130, each element of processor 140 directly or indirectly electrically connect between each other
It connects, to realize the transmission or interaction of data.For example, these elements can pass through one or more communication bus or signal between each other
Line, which is realized, to be electrically connected.The video recommendations device 110 includes at least one can be in the form of software or firmware (firmware)
It is stored in the memory 120 or is solidificated in the operating system (operating system, OS) of the electronic equipment 100
Software function module.The processor 140 is for executing the executable module stored in memory 120, such as the video
The software function module or computer program that recommendation apparatus 110 includes.
Wherein, memory 120 may be, but not limited to, random access memory (Random Access Memory,
RAM), read-only memory (Read Only Memory, ROM), programmable read only memory (Programmable Read-
Only Memory, PROM), erasable read-only memory (Erasable Programmable Read-Only Memory,
EPROM), electricallyerasable ROM (EEROM) (Electric Erasable Programmable Read-Only Memory,
EEPROM) etc..Wherein, memory 120 is for storing program, and the processor 140 is after receiving and executing instruction, described in execution
Program, method performed by the electronic equipment 100 for the flow definition that aftermentioned any embodiment of the embodiment of the present invention discloses can answer
It is realized in processor 140, or by processor 140.
Processor 140 may be a kind of IC chip, the processing capacity with signal.Above-mentioned processor can be
General processor, including central processing unit (Central Processing Unit, CPU), network processing unit (Network
Processor, NP) etc.;It can also be digital signal processor (DSP), specific integrated circuit (ASIC), field-programmable gate array
Arrange (FPGA) either other programmable logic device, discrete gate or transistor logic, discrete hardware components.It may be implemented
Or disclosed each method, step and logic diagram in the execution embodiment of the present invention.General processor can be microprocessor
Or the processor is also possible to any conventional processor etc..
Referring to Fig. 2, being a kind of video recommendations side applied to above-mentioned electronic equipment 100 provided in an embodiment of the present invention
Method, the step of including to it below in conjunction with Fig. 2, are illustrated.
Step S101: multiple video tabs are determined according to the consumption log of user, wherein the video tab is used for table
Show the classification of video.
The consumption log of user (such as user i) within a preset period of time (such as one month, 15 days, 7 days in the past) is obtained, and
Multiple video tabs are determined with this, namely determine corresponding video tab according to multiple videos of the customer consumption.Wherein,
Video tab is used to indicate the classification of video.It should be noted that each video has at least one video tab, the video
Label is to be divided in advance according to the content of video, and different video tabs indicates different types, which has a class
It is similar to the same to visual classification.
Wherein, consumption log includes: search log, plays log, click logs, download log, exposure log and do not like
Joyous log.Wherein, when consume log be play log, click logs, download log, expose log and do not like log when, it is right
What is answered is video itself, since each video is both provided with corresponding video tab in advance, when consumption log is to play day
Will, download log, when exposing log and not liking log, can directly determine out corresponding video tab at click logs.
But when consuming log is search log, corresponding video tab cannot be directly determined out, at this point it is possible to first to searching
Search term in Suo Zhi carries out word cutting processing as unit of word, obtains multiple words, then more to obtaining after word cutting processing again
A word carries out duplicate removal processing, then carries out permutation and combination to multiple words after duplicate removal, obtains multiple words, finally obtains permutation and combination
Multiple words in each word respectively with video library tag match, using with the word of video library tag hit as video tab.It is logical
Cross which, so that it may determine the corresponding video tab of search term.In order to make it easy to understand, citing is illustrated, it is assumed that the use
Within a preset period of time, a kind of to have searched for 3 times (truthful data be far above and this), search term difference is as follows: one goes out at family
Play (4 words), I be not medicine refreshing (5 words) and it is later we (5 words), after carrying out word cutting processing, altogether including 14 words,
After carrying out duplicate removal processing, 13 words are also remained, wherein eliminate duplicate " an I " word, then carry out to remaining 13 words
Permutation and combination includes 13 kinds of formulas such as in a manner of a combinatorics on words, includes 13*12=156 kind in a manner of two combinatorics on words, with
Triliteral combination includes 13*12*11=1716 ..., each word in multiple words for finally obtaining permutation and combination point
Not with video library tag match, using with the word of video library tag hit as video tab.
Step S102: the user is calculated to the preference value of each video tab in the multiple video tab.
After determining multiple video tabs according to the consumption log of user (such as user i), the user is calculated to multiple view
The preference value of each video tab in frequency marking label.
As an alternative embodiment, can be the quantity for counting each video tab, then divided by each of statistics
The sum of quantity of a video tab determines preference value.For example, determining 5 videos according to the consumption log of user (such as user i)
Label, respectively A, B, C, D and E, it is assumed that A statistics quantity be 5, B statistics quantity be 7, C statistics quantity be 10, D statistics
Quantity be 2 and the quantity of E statistics is 3, then user is 5/ (5+7+10+2+3) to the preference value of A, and similarly user is to the inclined of B
Good value is 7/ (5+7+10+2+3), and so on, preference value of the available user to each video tab.
As another optional embodiment, can be included in conjunction with Fig. 3 the step of, is illustrated the process.
Step S201: the last user calculated is obtained to each video tab in the multiple video tab
History preference value.
Since the history preference value of each video tab has record due to calculating before, in system, directly acquire i.e.
It can.If wherein some video tab is to calculate for the first time, which is defaulted as zero.
Step S202: the time interval at acquisition last calculating time point to current point in time, and according between the time
Every determining time attenuation coefficient.
The acquisition last time calculates time point to the time interval of current point in time, and determines the time according to the time interval
Attenuation coefficient, wherein the unit of the time interval is the second.For example, between last calculating time point to the time of current point in time
It is divided into 15 days, then the corresponding second is 15*24*60*60=1296000 seconds, determines that time attenuation coefficient can be with according to time interval
It is to determine that the time declines according to time attenuation function formula TimeDecay=exp (- deltaSeconds/decayFactor)
Subtract coefficient.Wherein, deltaSeconds is the number of seconds of last computation moment to current time apart, such as 1296000 seconds,
DecayFactor is for controlling rate of decay, for positive preference decayFactor=800000, for negative sense preference
DecayFactor=100000.
Wherein, it should be noted that positive preference and negative sense preference indicated is the type of video tab, i.e. video tab
If major class point comprising two classes, i.e., positive preference and negative sense preference, wherein be corresponding with multiple subclasses again under every class, such as
First-level class, secondary classification, content tab, LDA (Latent Dirichlet Allocation) label etc..If the video mark
Label belong to positive preference, then decayFactor=800000, if belonging to negative sense preference, decayFactor=100000.
Step S203: it calculates the user and all behaviors of each video tab in the multiple video tab is weighed
Weight.
According to each video tab, user is calculated separately to all behavior weights of each video tab, namely is clicked, broadcast
It puts, download, exposing, the sum of the weight for not liking five kinds of behaviors.It is to calculate user to all behavior weights of video tab j
Example, calculate separately first video tab j in click behavior, broadcasting behavior, downloading behavior, exposure behavior, do not like 5 kinds of rows of behavior
Then weight under can calculate user and weigh to all behaviors of video tab j to this 5 kinds of behavior weight summations again
Weight.In order to make it easy to understand, by taking following formula as an example: SumAllAction (ActionWeight*ItemWeight*
ItemTagWeight*UserTagWeight).Wherein, SumAllAction is all behavior weights comprising video tab j,
It include the sum of clicking, playing, downloading, exposing, not liking five kinds of behavior weights;ActionWeight is single behavior weight,
For positive preference, single behavior weight is respectively as follows: click behavior=1, plays behavior=2, downloads behavior=5, exposes behavior
=-0.1 does not like behavior=- 5, and for negative sense preference, single behavior weight is respectively as follows: click behavior=- 1, plays row
It is=- 2, downloads behavior=- 5, exposes behavior=1, do not like behavior=5.The power of ItemWeight expression video tab j
Weight, ItemWeight=log (ActionCount/ItemActionCount), wherein ActionCount indicates all user's needles
To the sum of some behavior of video tab j, ItemActionCount indicates user i for video tab j under the behavior
Sum.For example, then ActionCount indicates the total of click behavior of all users for video tab j by taking click behavior as an example
Number, it is assumed that comprising 100 users (wherein, user i is included within this 100 users), ActionCount is then this 100
User is directed to the sum of the hits of video tab j, it is assumed that is 10000.ItemActionCount indicates that user i is directed to video mark
Sign sum of the j under the behavior (being at this time click behavior), it is assumed that be 100.In another example by taking broadcasting behavior as an example, then
ActionCount indicates the sum of broadcasting behavior of all users for video tab j, it is assumed that (wherein, comprising 100 users
User i is included within this 100 users), ActionCount be then this 100 users for video tab j broadcasting number it
With, it is assumed that it is 80000.ItemActionCount indicates user i for video tab j the behavior (being at this time broadcasting behavior)
Under sum, it is assumed that be 500.ItemTagWeight indicates weight of the Tj in Item, is default value, usually 1, specific root
Factually depending on the setting value of border.UserTagWeight indicates weight of the Tj in user preference, UserTagWeight=log (institute
There is video tab history to tire out the number that calculated number/video tab j cumulative calculation is crossed).For example, still with above-mentioned 5
Video tab (A, B, C, D and E), for, then all video tab history tire out calculated number to be expressed as A, B, C, D and E each
The sum of calculated number is tired out from history.It is assumed that video tab j is A, then the number that video tab j cumulative calculation is crossed is then A
History tires out calculated number.
Step S204: institute is determined according to the history preference value, the time attenuation coefficient and all behavior weights
User is stated to the preference value of each video tab in the multiple video tab.
After the history preference value of each video tab of acquisition, time attenuation coefficient, all behavior weights, according to acquisition
History preference value, time attenuation coefficient and all behavior weights i.e. can determine user to each video in multiple video tabs
The preference value of label.
As an alternative embodiment, can be the history preference value of each video tab multiplied by time decaying system
Number adds the preference value that all behavior weights are the video tab.Its preference value of different calculation formula is different, can such as incite somebody to action
The history preference value of each video tab plus time attenuation coefficient is the preference of the video tab multiplied by all behavior weights
Value.
Wherein, in order to make it easy to understand, adding all behavior weights multiplied by time attenuation coefficient with history preference value is certain
It for the preference value of a video tab, is illustrated, according to the history preference value, the time attenuation coefficient and described all
Behavior weight and preference formula P (Ui, Tj)=P (Ui, Tj) ' * TimeDecay+SumAllAction (ActionWeight*
ItemWeight*ItemTagWeight*UserTagWeight) determine the user to each of the multiple video tab
The preference value of video tab.Wherein, P (Ui, Tj) indicates user i to the preference value of video tab j;P (Ui, Tj) ' indicates upper one
For secondary calculated user i to the history preference value of video tab j, calculating then value if first time is zero;When TimeDecay is indicated
Between attenuation coefficient=exp (- deltaSeconds/decayFactor), deltaSeconds indicates last and calculates time point
To the time interval of current point in time, unit is the second;DecayFactor is for controlling rate of decay, for positive preference
DecayFactor=800000, for negative sense preference decayFactor=100000;SumAllAction is comprising video mark
All behavior weights of j are signed, that is, include the sum of clicking, playing, downloading, exposing, not liking five kinds of behavior weights;
ActionWeight is single behavior weight, and for positive preference, single behavior weight is respectively as follows: click behavior=1, plays row
It is=2, downloads behavior=5, exposes behavior=- 0.1, do not like behavior=- 5, for negative sense preference, single behavior weight point
Not are as follows: click behavior=- 1, play behavior=- 2, download behavior=- 5, expose behavior=1, do not like behavior=5;
The weight of ItemWeight expression video tab j, ItemWeight=log (ActionCount/ItemActionCount),
Middle ActionCount indicates that the sum of some behavior of all users for video tab j, ItemActionCount indicate to use
Family i is directed to sum of the video tab j under the behavior;ItemTagWeight indicates weight of the Tj in Item, is default value;
UserTagWeight indicates weight of the Tj in user preference, and (all video tab history are accumulative by UserTagWeight=log
The number that the number calculated/video tab j cumulative calculation is crossed).
Step S103: it is determined according to the preference value of each video tab and recommends video, and recommended to client.
After calculating user to the preference value of each video tab, according to the preference value of calculated each video tab
It determines and recommends video, and recommend to client.The video recommended from the preference thus according to user to it, it is ensured that can be user
Recommend its interested video.It is alternatively possible to be to promote decision tree (Gradient Boosted using trained gradient
Decision Tree, GBDT) model is ranked up the preference value of each video tab, ranking results are obtained, further according to
The ranking results determine the recommendation video.For example, directly choosing result conduct in the top after obtaining ranking results and pushing away
Video is recommended, if ranking first three content is as consequently recommended content.Wherein, preference of the GBDT in addition to considering user mentioned above
Feature (namely user is to preference value of each video tab) outside, also by the way that long to play time and video is complete broadcasts rate
Relatively high samples played assigns higher weight, and final ranking results is enabled preferably to reflect the preference of user.
Wherein, it should be noted that it is that precondition is good that the gradient, which promotes decision-tree model, comprising three processes,
First, training log;Second, sample characteristics extract, third, export result optimizing, and the process and the existing gradient promote decision
The establishment process of tree-model is similar, unlike, trained data are different, so that application scenarios are different.
Wherein, result optimizing is exported, that is, in addition to (namely user is to every for the preference profiles that consider user mentioned above
The preference value of a video tab) outside, also by the way that long to play time and video is complete broadcasts the relatively high samples played of rate
Higher weight is assigned, final ranking results is enabled preferably to reflect the preference of user.
That is, also then to be carried out in conjunction with corresponding weight in the preference value of calculated each video tab
Sequence, this recommendation video determined by way of multiple target weight optimization, compared to being based only on each video tab
The recommendation video determined of preference value, can more reflect the interest of user, and then consequently recommended video can more attract user.
It should be noted that the above-mentioned example shown be in order to make it easy to understand, and for example, therefore, can not will
It is understood as being the limitation to the application.
The embodiment of the present invention additionally provides a kind of video recommendations device 110, as shown in Figure 4.The video recommendations device 110
It include: the first determining module 111, computing module 112 and the second determining module 113.
First determining module 111, for determining multiple video tabs according to the consumption log of user, wherein the view
Frequency marking label are used to indicate the classification of video.
Computing module 112, for calculating the user to the preference of each video tab in the multiple video tab
Value,
Second determines mould 113, recommends video for determining according to the preference value of each video tab, and recommend to visitor
Family end.
It should be noted that all the embodiments in this specification are described in a progressive manner, each embodiment weight
Point explanation is the difference from other embodiments, and the same or similar parts between the embodiments can be referred to each other.
The technical effect of video recommendations device 110 provided by the embodiment of the present invention, realization principle and generation and aforementioned
Embodiment of the method is identical, and to briefly describe, Installation practice part does not refer to place, can refer to corresponding in preceding method embodiment
Content.
The embodiment of the invention also provides a kind of non-volatile computer read/write memory medium, which includes meter
Calculation machine program, the computer program execute above-mentioned method when being run by processor.
In several embodiments provided herein, it should be understood that disclosed device and method can also pass through
Other modes are realized.The apparatus embodiments described above are merely exemplary, for example, flow chart and block diagram in attached drawing
Show the device of multiple embodiments according to the present invention, the architectural framework in the cards of method and computer program product,
Function and operation.In this regard, each box in flowchart or block diagram can represent the one of a module, section or code
Part, a part of the module, section or code, which includes that one or more is for implementing the specified logical function, to be held
Row instruction.It should also be noted that function marked in the box can also be to be different from some implementations as replacement
The sequence marked in attached drawing occurs.For example, two continuous boxes can actually be basically executed in parallel, they are sometimes
It can execute in the opposite order, this depends on the function involved.It is also noted that every in block diagram and or flow chart
The combination of box in a box and block diagram and or flow chart can use the dedicated base for executing defined function or movement
It realizes, or can realize using a combination of dedicated hardware and computer instructions in the system of hardware.
In addition, each functional module in each embodiment of the present invention can integrate one independent portion of formation together
Point, it is also possible to modules individualism, an independent part can also be integrated to form with two or more modules.
It, can be with if the function is realized and when sold or used as an independent product in the form of software function module
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, laptop, server or network equipment etc.) execute the whole of each embodiment the method for the present invention
Or part steps.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.It should be noted that, in this document, relational terms such as first and second and the like are used merely to one
A entity or operation with another entity or operate distinguish, without necessarily requiring or implying these entities or operation it
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant are intended to
Cover non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or setting
Standby intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in the process, method, article or apparatus that includes the element.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of video recommendation method characterized by comprising
Multiple video tabs are determined according to the consumption log of user, wherein the video tab is used to indicate the classification of video;
The user is calculated to the preference value of each video tab in the multiple video tab;
It is determined according to the preference value of each video tab and recommends video, and recommended to client.
2. the method according to claim 1, wherein calculating the user to every in the multiple video tab
The preference value of a video tab, comprising:
The last user calculated is obtained to the history preference value of each video tab in the multiple video tab;
The acquisition last time calculates time point to the time interval of current point in time, and determines that the time decays according to the time interval
Coefficient;
The user is calculated to all behavior weights of each video tab in the multiple video tab;
Determine the user to described more according to the history preference value, the time attenuation coefficient and all behavior weights
The preference value of each video tab in a video tab.
3. according to the method described in claim 2, it is characterized in that, according to the history preference value, the time attenuation coefficient
Determine the user to the preference value of each video tab in the multiple video tab, packet with all behavior weights
It includes:
According to the history preference value, the time attenuation coefficient and all behavior weights and preference formula P (Ui, Tj)
=P (Ui, Tj) ' * TimeDecay+SumAllAction (ActionWeight*ItemWeight*ItemTagWeight*
UserTagWeight) determine the user to the preference value of each video tab in the multiple video tab;
Wherein, P (Ui, Tj) indicates user i to the preference value of video tab j;P (Ui, Tj) ' indicates last calculated user
For i to the history preference value of video tab j, calculating then value if first time is zero;TimeDecay expression time attenuation coefficient=
Exp (- deltaSeconds/decayFactor), deltaSeconds indicate last and calculate time point to current point in time
Time interval, unit is the second;DecayFactor is for controlling rate of decay, for positive preference decayFactor=
800000, for negative sense preference decayFactor=100000;SumAllAction is all behaviors comprising video tab j
Weight includes the sum of clicking, playing, downloading, exposing, not liking five kinds of behavior weights;ActionWeight is single behavior
Weight, for positive preference, single behavior weight is respectively as follows: click behavior=1, plays behavior=2, downloads behavior=5, exposure
Behavior=- 0.1 does not like behavior=- 5, and for negative sense preference, single behavior weight is respectively as follows: click behavior=- 1, broadcasts
Letting pass is=- 2, downloads behavior=- 5, exposes behavior=1, do not like behavior=5;ItemWeight indicates video tab j's
Weight, ItemWeight=log (ActionCount/ItemActionCount), wherein ActionCount indicates all users
For the sum of some behavior of video tab j, ItemActionCount indicates user i for video tab j under the behavior
Sum;ItemTagWeight indicates weight of the Tj in Item, is default value;UserTagWeight indicates that Tj is inclined in user
Weight in good, (all video tab history tire out the accumulative meter of calculated number/video tab j to UserTagWeight=log
The number calculated).
4. the method according to claim 1, wherein the consumption log is search log;According to disappearing for user
Multiple video tabs are determined in expense log, comprising:
Word cutting processing is carried out as unit of word to the search term in described search log, obtains multiple words;
Duplicate removal processing is carried out to the multiple word;
Permutation and combination is carried out to multiple words after duplicate removal, obtains multiple words;
By each word in the multiple word respectively with video library tag match, using with the word of the video library tag hit as
The video tab.
5. recommending the method according to claim 1, wherein being determined according to the preference value of each video tab
Video, comprising:
Decision-tree model is promoted using trained gradient to be ranked up the preference value of each video tab, is sorted
As a result;
The recommendation video is determined according to the ranking results.
6. a kind of video recommendations device characterized by comprising
First determining module, for determining multiple video tabs according to the consumption log of user, wherein the video tab is used
In the classification for indicating video;
Computing module, for calculating the user to the preference value of each video tab in the multiple video tab;
Second determining module is recommended video for determining according to the preference value of each video tab, and is recommended to client.
7. device according to claim 6, which is characterized in that the computing module is also used to obtain last calculating
History preference value of the user to each video tab in the multiple video tab;Acquisition last calculating time point arrives
The time interval of current point in time, and time attenuation coefficient is determined according to the time interval;The user is calculated to described more
All behavior weights of each video tab in a video tab;According to the history preference value, the time attenuation coefficient
Determine the user to the preference value of each video tab in the multiple video tab with all behavior weights.
8. device according to claim 7, which is characterized in that the computing module is also used to according to the history preference
Value, the time attenuation coefficient and all behavior weights and preference formula P (Ui, Tj)=P (Ui, Tj) ' *
TimeDecay+SumAllAction (ActionWeight*ItemWeight*ItemTagWeight*
UserTagWeight) determine the user to the preference value of each video tab in the multiple video tab;
Wherein, P (Ui, Tj) indicates user i to the preference value of video tab j;P (Ui, Tj) ' indicates last calculated user
For i to the history preference value of video tab j, calculating then value if first time is zero;TimeDecay expression time attenuation coefficient=
Exp (- deltaSeconds/decayFactor), deltaSeconds indicate last and calculate time point to current point in time
Time interval, unit is the second;DecayFactor is for controlling rate of decay, for positive preference decayFactor=
800000, for negative sense preference decayFactor=100000;SumAllAction is all behaviors comprising video tab j
Weight includes the sum of clicking, playing, downloading, exposing, not liking five kinds of behavior weights;ActionWeight is single behavior
Weight, for positive preference, single behavior weight is respectively as follows: click behavior=1, plays behavior=2, downloads behavior=5, exposure
Behavior=- 0.1 does not like behavior=- 5, and for negative sense preference, single behavior weight is respectively as follows: click behavior=- 1, broadcasts
Letting pass is=- 2, downloads behavior=- 5, exposes behavior=1, do not like behavior=5;ItemWeight indicates video tab j's
Weight, ItemWeight=log (ActionCount/ItemActionCount), wherein ActionCount indicates all users
For the sum of some behavior of video tab j, ItemActionCount indicates user i for video tab j under the behavior
Sum;ItemTagWeight indicates weight of the Tj in Item, is default value;UserTagWeight indicates that Tj is inclined in user
Weight in good, (all video tab history tire out the accumulative meter of calculated number/video tab j to UserTagWeight=log
The number calculated).
9. a kind of electronic equipment characterized by comprising memory and processor, the memory are connected with the processor;
The memory, for storing program;
The processor, for calling the program being stored in the memory, to execute such as any one of claim 1-5
The method.
10. a kind of storage medium, which is characterized in that including computer program, the computer program is held when being run by processor
Row method according to any one of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811507325.3A CN109657138B (en) | 2018-12-10 | 2018-12-10 | Video recommendation method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811507325.3A CN109657138B (en) | 2018-12-10 | 2018-12-10 | Video recommendation method and device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109657138A true CN109657138A (en) | 2019-04-19 |
CN109657138B CN109657138B (en) | 2021-02-26 |
Family
ID=66113293
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811507325.3A Active CN109657138B (en) | 2018-12-10 | 2018-12-10 | Video recommendation method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109657138B (en) |
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110209843A (en) * | 2019-05-31 | 2019-09-06 | 腾讯科技(深圳)有限公司 | Multimedia resource playback method, device, equipment and storage medium |
CN110245274A (en) * | 2019-04-23 | 2019-09-17 | 五八有限公司 | A kind of label temperature calculates method, apparatus, electronic equipment and storage medium |
CN110287372A (en) * | 2019-06-26 | 2019-09-27 | 广州市百果园信息技术有限公司 | Label for negative-feedback determines method, video recommendation method and its device |
CN110321474A (en) * | 2019-05-21 | 2019-10-11 | 北京奇艺世纪科技有限公司 | Recommended method, device, terminal device and storage medium based on search term |
CN110413881A (en) * | 2019-07-11 | 2019-11-05 | 腾讯科技(深圳)有限公司 | A kind of method, apparatus, the network equipment and storage medium identifying label accuracy |
CN110929166A (en) * | 2019-12-27 | 2020-03-27 | 咪咕文化科技有限公司 | Content recommendation method, electronic device and storage medium |
CN111125514A (en) * | 2019-11-20 | 2020-05-08 | 泰康保险集团股份有限公司 | User behavior analysis method and device, electronic equipment and storage medium |
CN111159541A (en) * | 2019-12-11 | 2020-05-15 | 微民保险代理有限公司 | Method and device for determining account behavior preference |
CN111209432A (en) * | 2020-01-02 | 2020-05-29 | 北京字节跳动网络技术有限公司 | Information acquisition method and device, electronic equipment and computer readable medium |
CN111510783A (en) * | 2020-04-26 | 2020-08-07 | 咪咕动漫有限公司 | Method, device, electronic equipment and storage medium for determining video exposure |
CN111581452A (en) * | 2020-03-26 | 2020-08-25 | 浙江口碑网络技术有限公司 | Method and device for obtaining recommendation object data and electronic equipment |
CN111666450A (en) * | 2020-06-04 | 2020-09-15 | 北京奇艺世纪科技有限公司 | Video recall method and device, electronic equipment and computer-readable storage medium |
CN111738768A (en) * | 2020-06-24 | 2020-10-02 | 江苏云柜网络技术有限公司 | Advertisement pushing method and system |
CN111813992A (en) * | 2020-07-14 | 2020-10-23 | 四川长虹电器股份有限公司 | Sorting system and method for movie recommendation candidate set |
CN112135193A (en) * | 2020-09-24 | 2020-12-25 | 湖南快乐阳光互动娱乐传媒有限公司 | Video recommendation method and device |
WO2021052041A1 (en) * | 2019-09-20 | 2021-03-25 | 北京字节跳动网络技术有限公司 | Video pushing method and apparatus based on video search, and electronic device |
CN113139083A (en) * | 2020-01-19 | 2021-07-20 | Tcl集团股份有限公司 | Video recommendation method and device, terminal equipment and storage medium |
CN113407521A (en) * | 2021-05-24 | 2021-09-17 | 广州市万表科技股份有限公司 | User behavior tag preference sorting method, device, equipment and storage medium |
CN114048392A (en) * | 2022-01-13 | 2022-02-15 | 北京达佳互联信息技术有限公司 | Multimedia resource pushing method and device, electronic equipment and storage medium |
WO2022068492A1 (en) * | 2020-09-29 | 2022-04-07 | 百果园技术(新加坡)有限公司 | Video recommendation method and apparatus |
CN114528484A (en) * | 2022-01-26 | 2022-05-24 | 北京金堤科技有限公司 | Preference mining method and device, storage medium and electronic equipment |
CN117194772A (en) * | 2023-08-17 | 2023-12-08 | 广州兴趣岛信息科技有限公司 | Content pushing method and device based on user tag |
CN111666450B (en) * | 2020-06-04 | 2024-04-26 | 北京奇艺世纪科技有限公司 | Video recall method, device, electronic equipment and computer readable storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104602042A (en) * | 2014-12-31 | 2015-05-06 | 合一网络技术(北京)有限公司 | User behavior based label setting method |
CN104933134A (en) * | 2015-06-12 | 2015-09-23 | 海信集团有限公司 | User feature analysis method and user feature analysis device |
CN105005587A (en) * | 2015-06-26 | 2015-10-28 | 深圳市腾讯计算机系统有限公司 | User portrait updating method, apparatus and system |
CN106407241A (en) * | 2016-03-21 | 2017-02-15 | 传线网络科技(上海)有限公司 | Video recommendation method and system |
-
2018
- 2018-12-10 CN CN201811507325.3A patent/CN109657138B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104602042A (en) * | 2014-12-31 | 2015-05-06 | 合一网络技术(北京)有限公司 | User behavior based label setting method |
CN104933134A (en) * | 2015-06-12 | 2015-09-23 | 海信集团有限公司 | User feature analysis method and user feature analysis device |
CN105005587A (en) * | 2015-06-26 | 2015-10-28 | 深圳市腾讯计算机系统有限公司 | User portrait updating method, apparatus and system |
CN106407241A (en) * | 2016-03-21 | 2017-02-15 | 传线网络科技(上海)有限公司 | Video recommendation method and system |
Cited By (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110245274A (en) * | 2019-04-23 | 2019-09-17 | 五八有限公司 | A kind of label temperature calculates method, apparatus, electronic equipment and storage medium |
CN110321474A (en) * | 2019-05-21 | 2019-10-11 | 北京奇艺世纪科技有限公司 | Recommended method, device, terminal device and storage medium based on search term |
CN110321474B (en) * | 2019-05-21 | 2022-01-11 | 北京奇艺世纪科技有限公司 | Recommendation method and device based on search terms, terminal equipment and storage medium |
CN110209843A (en) * | 2019-05-31 | 2019-09-06 | 腾讯科技(深圳)有限公司 | Multimedia resource playback method, device, equipment and storage medium |
CN110287372A (en) * | 2019-06-26 | 2019-09-27 | 广州市百果园信息技术有限公司 | Label for negative-feedback determines method, video recommendation method and its device |
CN110287372B (en) * | 2019-06-26 | 2021-06-01 | 广州市百果园信息技术有限公司 | Label determining method for negative feedback, video recommending method and device thereof |
CN110413881B (en) * | 2019-07-11 | 2023-10-20 | 腾讯科技(深圳)有限公司 | Method, device, network equipment and storage medium for identifying label accuracy |
CN110413881A (en) * | 2019-07-11 | 2019-11-05 | 腾讯科技(深圳)有限公司 | A kind of method, apparatus, the network equipment and storage medium identifying label accuracy |
WO2021052041A1 (en) * | 2019-09-20 | 2021-03-25 | 北京字节跳动网络技术有限公司 | Video pushing method and apparatus based on video search, and electronic device |
CN111125514A (en) * | 2019-11-20 | 2020-05-08 | 泰康保险集团股份有限公司 | User behavior analysis method and device, electronic equipment and storage medium |
CN111125514B (en) * | 2019-11-20 | 2023-08-22 | 泰康保险集团股份有限公司 | Method, device, electronic equipment and storage medium for analyzing user behaviors |
CN111159541B (en) * | 2019-12-11 | 2023-08-25 | 微民保险代理有限公司 | Method and device for determining account behavior preference |
CN111159541A (en) * | 2019-12-11 | 2020-05-15 | 微民保险代理有限公司 | Method and device for determining account behavior preference |
CN110929166B (en) * | 2019-12-27 | 2023-10-20 | 咪咕文化科技有限公司 | Content recommendation method, electronic equipment and storage medium |
CN110929166A (en) * | 2019-12-27 | 2020-03-27 | 咪咕文化科技有限公司 | Content recommendation method, electronic device and storage medium |
CN111209432A (en) * | 2020-01-02 | 2020-05-29 | 北京字节跳动网络技术有限公司 | Information acquisition method and device, electronic equipment and computer readable medium |
CN113139083A (en) * | 2020-01-19 | 2021-07-20 | Tcl集团股份有限公司 | Video recommendation method and device, terminal equipment and storage medium |
CN111581452B (en) * | 2020-03-26 | 2023-10-17 | 浙江口碑网络技术有限公司 | Recommendation object data obtaining method and device and electronic equipment |
CN111581452A (en) * | 2020-03-26 | 2020-08-25 | 浙江口碑网络技术有限公司 | Method and device for obtaining recommendation object data and electronic equipment |
CN111510783B (en) * | 2020-04-26 | 2022-06-03 | 咪咕动漫有限公司 | Method, device, electronic equipment and storage medium for determining video exposure |
CN111510783A (en) * | 2020-04-26 | 2020-08-07 | 咪咕动漫有限公司 | Method, device, electronic equipment and storage medium for determining video exposure |
CN111666450A (en) * | 2020-06-04 | 2020-09-15 | 北京奇艺世纪科技有限公司 | Video recall method and device, electronic equipment and computer-readable storage medium |
CN111666450B (en) * | 2020-06-04 | 2024-04-26 | 北京奇艺世纪科技有限公司 | Video recall method, device, electronic equipment and computer readable storage medium |
CN111738768A (en) * | 2020-06-24 | 2020-10-02 | 江苏云柜网络技术有限公司 | Advertisement pushing method and system |
CN111813992A (en) * | 2020-07-14 | 2020-10-23 | 四川长虹电器股份有限公司 | Sorting system and method for movie recommendation candidate set |
CN112135193A (en) * | 2020-09-24 | 2020-12-25 | 湖南快乐阳光互动娱乐传媒有限公司 | Video recommendation method and device |
WO2022068492A1 (en) * | 2020-09-29 | 2022-04-07 | 百果园技术(新加坡)有限公司 | Video recommendation method and apparatus |
CN113407521B (en) * | 2021-05-24 | 2022-02-08 | 广州市万表科技股份有限公司 | User behavior tag preference sorting method, device, equipment and storage medium |
CN113407521A (en) * | 2021-05-24 | 2021-09-17 | 广州市万表科技股份有限公司 | User behavior tag preference sorting method, device, equipment and storage medium |
CN114048392A (en) * | 2022-01-13 | 2022-02-15 | 北京达佳互联信息技术有限公司 | Multimedia resource pushing method and device, electronic equipment and storage medium |
CN114528484A (en) * | 2022-01-26 | 2022-05-24 | 北京金堤科技有限公司 | Preference mining method and device, storage medium and electronic equipment |
CN117194772A (en) * | 2023-08-17 | 2023-12-08 | 广州兴趣岛信息科技有限公司 | Content pushing method and device based on user tag |
CN117194772B (en) * | 2023-08-17 | 2024-04-30 | 广州兴趣岛信息科技有限公司 | Content pushing method and device based on user tag |
Also Published As
Publication number | Publication date |
---|---|
CN109657138B (en) | 2021-02-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109657138A (en) | A kind of video recommendation method, device, electronic equipment and storage medium | |
US10824682B2 (en) | Enhanced online user-interaction tracking and document rendition | |
CN105045831B (en) | A kind of information push method and device | |
CN104065565B (en) | The method of PUSH message, server, client terminal device and system | |
CN110413877A (en) | A kind of resource recommendation method, device and electronic equipment | |
CN110532451A (en) | Search method and device for policy text, storage medium, electronic device | |
CN102811371B (en) | The method, system and device that intelligent television application program is recommended | |
CN109582857A (en) | Based on big data information-pushing method, device, computer equipment and storage medium | |
US20090292677A1 (en) | Integrated web analytics and actionable workbench tools for search engine optimization and marketing | |
CA3153598A1 (en) | Method of and device for predicting video playback integrity | |
CN106940705A (en) | A kind of method and apparatus for being used to build user's portrait | |
CN107689008A (en) | A kind of user insures the method and device of behavior prediction | |
US20090299998A1 (en) | Keyword discovery tools for populating a private keyword database | |
CN110222975A (en) | A kind of loss customer analysis method, apparatus, electronic equipment and storage medium | |
CN109511015B (en) | Multimedia resource recommendation method, device, storage medium and equipment | |
US20160285672A1 (en) | Method and system for processing network media information | |
US11068926B2 (en) | System and method for analyzing and predicting emotion reaction | |
CN104462594A (en) | Method and device for providing user personalized resource message pushing | |
CN107093091B (en) | Data processing method and device | |
CN109543132A (en) | Content recommendation method, device, electronic equipment and storage medium | |
CN108363730B (en) | Content recommendation method, system and terminal equipment | |
CN110069676A (en) | Keyword recommendation method and device | |
CN110689402A (en) | Method and device for recommending merchants, electronic equipment and readable storage medium | |
CN110706015A (en) | Advertisement click rate prediction oriented feature selection method | |
CN114398560B (en) | Marketing interface setting method, device, equipment and medium based on WEB platform |
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 | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20210203 Address after: Unit 1701e, China energy storage building, 3099 Keyuan South Road, high tech community, Yuehai street, Nanshan District, Shenzhen, Guangdong 518000 Applicant after: SHENZHEN DAYU WUXIAN TECHNOLOGY Co.,Ltd. Address before: Unit 2301-l, bicker building, No.9, Keke Road, Gaoxin Middle District, Yuehai street, Nanshan District, Shenzhen, Guangdong 518000 Applicant before: SHENZHEN MOSHI TECHNOLOGY Co.,Ltd. |
|
TA01 | Transfer of patent application right | ||
GR01 | Patent grant | ||
GR01 | Patent grant |