CN103763361B - A kind of method, system and recommendation server for recommending application based on user behavior - Google Patents
A kind of method, system and recommendation server for recommending application based on user behavior Download PDFInfo
- Publication number
- CN103763361B CN103763361B CN201410014009.8A CN201410014009A CN103763361B CN 103763361 B CN103763361 B CN 103763361B CN 201410014009 A CN201410014009 A CN 201410014009A CN 103763361 B CN103763361 B CN 103763361B
- Authority
- CN
- China
- Prior art keywords
- application
- recommendation
- identifier
- download
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000006243 chemical reaction Methods 0.000 claims abstract description 81
- 230000004044 response Effects 0.000 claims description 28
- 238000001914 filtration Methods 0.000 claims description 23
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000004422 calculation algorithm Methods 0.000 description 12
- 238000009434 installation Methods 0.000 description 12
- 238000005516 engineering process Methods 0.000 description 5
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000011524 similarity measure Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013501 data transformation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a kind of method, system and recommendation server for recommending application based on user behavior, wherein method includes：The first identifier applied for obtaining the identifier for the first application that user browses and downloading, and the identifier with second application of described first application with recommendation relation；The number of visits and download time of the application of statistics first；According to the download time and number of visits of first application of statistics, the download conversion ratio that first application is downloaded according to the recommendation relation of the second application and the first application is calculated.By downloading conversion ratio can guide user's which application of selection download from the multiple first applications according to the download conversion ratio of the first application, so as to improve the success rate that the first application is recommended to download.
Description
Technical field
The invention belongs to field of computer technology, more particularly to a kind of method, system for recommending application based on user behavior
And recommendation server.
Background technology
As the continuous development of terminal technology, such as the terminal such as smart mobile phone, tablet computer have stronger and stronger place
Reason ability, user can be by installing various application programs, to realize a variety of functions in terminal.In order to meet user
, there are more and more application programs, therefore in various demands, on the one hand,
For application program, how to find oneself desired application program also becomes problem to be solved.
The content of the invention
In view of the above problems, it is proposed that the present invention overcomes the above problem in order to provide one kind or solves at least in part
State method, system and the recommendation server of recommending application based on user behavior of problem.
According to one aspect of the present invention, there is provided a kind of method for recommending application based on user behavior, including：
The identifier of the first application for obtaining the identifier for the first application that user browses and downloading, and with described first
Using the identifier with the second application for recommending relation；
Count the number of visits and download time of first application；
According to the download time and number of visits of first application of statistics, calculating should according to the second application and first
Relation is recommended to download the download conversion ratio of first application.
Alternatively, before download conversion ratio is calculated, the method further includes：
According to the recommendation relation between the described first application and the second application, the identifier and the of first application is preserved
The correspondence of the identifier of two applications, and the number of visits and download time of the first application.
Alternatively, the recommendation relation between first application and the second application is given birth to by the collaborative filtering based on article
Into.
Alternatively, the method further includes：
The behavior of the first application is browsed in response to user, user is recorded in the daily record of the recommendation application of network side server
The identifier of current the first application browsed, and the mark with second application of described first application with recommendation relation
Symbol.
Alternatively, the method further includes：
The behavior of first application is downloaded in response to user, in the daily record of the recommendation application of the network side server
The identifier for the first application that record user is downloading, and applied with second of the described first application with recommendation relation
Identifier.
Alternatively, the identifier for obtaining the first application that user browses and the identifier for the first application downloaded, with
And before the step of identifier of the second application with recommendation relation is applied with described first, the method further includes：
Judge whether the daily record in the network side server is the daily record for recommending application one by one；
If recommending the daily record of application, then enter and obtain the browsed identifier with the downloaded first application of user, with
And there is the step of identifier of the second application of recommendation relation with the described first application.
Alternatively, the method further includes：
The request of second application is browsed or downloaded in response to user, and generating application in recommendation server side recommends page
Face, the application displayed page include：Using region and application recommendation region is downloaded, include wherein region is downloaded in application：The
The description information of two applications, includes using region is recommended：Applied with the second one or more first of the application with recommendation relation
Description information, and it is each first application download conversion ratio；
The application displayed page is sent to terminal device in the recommendation application server side, and is set in the terminal
Standby upper display.
Alternatively, when it is described using have on displayed page it is multiple first application description information when, the multiple first should
Description information is arranged in order according to the size of the download conversion ratio of the described first application.
Alternatively, the description information of first application includes：The identifier of first application, the icon of the first application, the
The download time of the download address of one application, the number of visits of the first application and the first application；
The description information of second application includes：The identifier of second application, the icon of the second application and the second application
Download address.
Alternatively, the description information of first application further includes：The rationale for the recommendation of first application.
According to another aspect of the present invention, a kind of recommendation server is also provided, including：
Acquisition module, the identifier of the first application browsed for obtaining user and the identifier for the first application downloaded,
And the identifier with second application of described first application with recommendation relation；
Statistical module, for counting the number of visits and download time of first application；
Conversion ratio generation module is downloaded, for the download time and number of visits of first application according to statistics, meter
Calculate the download conversion ratio that first application is downloaded according to the recommendation relation of the second application and the first application.
Alternatively, the recommendation server further includes：
Preserving module, for according to the recommendation relation between the described first application and the second application, preserving described first should
The correspondence of identifier and the identifier of the second application, and the number of visits and download time of the first application.
Alternatively, the recommendation relation between first application and the second application is given birth to by the collaborative filtering based on article
Into.
Alternatively, the recommendation server further includes：
Log analysis module, for judge the daily record whether be recommend application daily record；If recommending the daily record of application,
Then trigger the acquisition module.
Alternatively, the recommendation server further includes：
Logging modle is browsed, for browsing the behavior of the first application in response to user, is recorded in the daily record for recommending application
The identifier for the first application that user is currently browsing, and applied with second of the described first application with recommendation relation
Identifier.
Alternatively, the recommendation server further includes：
Download History module, for downloading the behavior of first application in response to user, in the daily record for recommending application
The identifier for the first application that record user is downloading, and applied with second of the described first application with recommendation relation
Identifier.
Alternatively, the recommendation server further includes：
Using page generation module is recommended, for the request of second application, generation to be browsed or downloaded in response to user
Using the page is recommended, the application displayed page includes：Using region and application recommendation region is downloaded, wherein region is downloaded in application
Include：The description information of second application, includes using region is recommended：With the second one or more of the application with recommendation relation
The description information of first application, and the download conversion ratio of each first application；
Using page sending module is recommended, for the application displayed page to be sent to terminal device, and at the end
Shown in end equipment.
Alternatively, it is described using on displayed page include it is multiple it is described first application identifiers, the multiple first
The identifier of application is arranged in order according to the size of the download conversion ratio of the described first application.
Alternatively, the description information of first application includes：The identifier of first application, the icon of the first application, the
The download time of the download address of one application, the number of visits of the first application and the first application；
The description information of second application includes：The identifier of second application, the icon of the second application and the second application
Download address.
Alternatively, the description information of first application further includes：The rationale for the recommendation of first application.
According to an additional aspect of the present invention, a kind of system for recommending application based on user behavior is additionally provided, including such as
The upper recommendation server.
As shown from the above technical solution, the embodiment of the present invention has the advantages that：First is browsed in response to user
Using and download first application behavior, record browse first application identifier（ID）The first application downloaded with record
Identifier（ID）, and the identifier with second application of first application with recommendation relation（ID）, then counting first should
Number of visits and download time, and the download of the first application is calculated according to the number of visits and download time of the first application
Conversion ratio, the download conversion ratio of first application can reflect user after the first application is browsed, and successfully downloading first should
Probability, the bigger explanation user of probability may more download the first application.When user request browse or download second in application,
In response to the request of user, can be generated using displayed page in recommendation server side, this includes using displayed page：Under
Carry region and region is recommended in application, can include in region wherein application is downloaded：The identifier of second application, the figure of the second application
Mark and second application download address, using recommend region in can include：With the second application with one of recommendation relation or
The identifier of multiple first applications, the download address of each first application, and the download conversion ratio of each first application, due to
User can be reflected after the first application is browsed by downloading conversion ratio, the probability of the first application successfully be downloaded, so as to guide use
Family selects to download according to the conversion ratio of downloading of the first application from multiple first applications of recommendation, such as selection downloads conversion ratio most
The first big application, so as to improve the success rate that the first application is recommended to download, and due to download conversion ratio take into account it is each
The navigation patterns of user and behavior is downloaded, so as to fulfill the difference for the behavior that can combine each user, to the user accurately
Recommend the application for meeting the user's demand.
Brief description of the drawings
By reading the detailed description of hereafter preferred embodiment, it is various other the advantages of and benefit it is common for this area
Technical staff will be clear understanding.Attached drawing is only used for showing the purpose of preferred embodiment, and is not considered as to the present invention
Limitation.And in whole attached drawing, identical component is denoted by the same reference numerals.In the accompanying drawings：
Fig. 1 recommends one of flow chart of method of application based on user behavior in showing according to an embodiment of the invention；
Fig. 2 recommends the two of the flow chart of the method for application based on user behavior in showing according to an embodiment of the invention；
Fig. 3 shows the block diagram of middle recommendation server according to an embodiment of the invention；
Fig. 4 recommends the block diagram of the system of application based on user behavior in showing according to an embodiment of the invention；And
Fig. 5 recommends the block schematic illustration of the system of application based on user behavior in showing according to an embodiment of the invention.
Embodiment
The exemplary embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although the disclosure is shown in attached drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
Limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
Completely it is communicated to those skilled in the art.
At present, for application, occur and apply relevant ranking list, such as：Based on application scoring ranking list or
The ranking list of human-edited is carried out to application according to advertising expense.Although these ranking lists select application to provide necessarily for user
Reference.But this mode for recommending application is directed to all users, and the difference of the behavior due to not accounting for each user, leads
The application for causing to recommend cannot accurately meet the demand of each user.
For this problem, the inventors discovered that the first application can be browsed in response to user and download the row of the first application
To record the first identifier applied browsed（ID）The identifier for the first application downloaded with record（ID）, and with this
The identifier of second application of one application with recommendation relation（ID）, then count the number of visits of the first application and download secondary
Count, and the download conversion ratio of the first application is calculated according to the number of visits and download time of the first application, first application
User can be reflected after the first application is browsed by downloading conversion ratio, successfully download the probability of the first application, the bigger theory of probability
Bright user more may download the first application.When user's request browses or downloads second in application, request in response to user,
Recommendation server side can be generated using displayed page, this includes using displayed page：Using download region and using recommended area
Domain, can include wherein application is downloaded in region：The download of the identifier of second application, the icon of the second application and the second application
Address, using recommend region in can include：The mark applied with the second one or more first of the application with recommendation relation
Symbol, the download address of each first application, and the download conversion ratio of each first application, can reflect due to downloading conversion ratio
Go out user after the first application is browsed, successfully download the probability of the first application, so as to guide user according under the first application
Idling rate selects to download from multiple first applications of recommendation, such as the first maximum application of conversion ratio is downloaded in selection, so that
The success rate downloaded is recommended in the application of raising first, and take into account the navigation patterns of each user with due to downloading conversion ratio
Load behavior, so as to fulfill the difference for the behavior that can combine each user, accurately recommends to meet the user's demand to the user
Application.
After the basic principle of the present invention is described, lower mask body introduces the various non-limiting embodiment party of the present invention
Formula.
As shown in Figure 1, to recommend one of flow chart of method of application in the embodiment of the present invention based on user behavior, should
The executive agent of method is recommendation server, and this method 100 includes：
Step S101, the identifier for the first application that user browses and the identifier of the first application of download are obtained, and
With the identifier of second application of first application with recommendation relation.
In an embodiment of the present invention, the first application and the second application refer to possess unique identifier（Such as digital sequence
Row number）Software, which can include the software suitable for computing device and the software suitable for mobile equipment.
In the embodiment of the present invention, the navigation patterns for browsing the first application that can be based on user and the application of download first
Download behavior, in network side server（Such as log server or recommendation server）The first application that middle record user browses
Identifier and download first application identifier.Wherein, user browses the application displaying of the first application on the terminal device
The situation of the page, it will be appreciated that browse first application for user, this believes using the description that record on displayed page has the first application
Breath, the description information include：The identifier of first application, the icon of the first application, the download address of the first application；User clicks on
Download the situation of the download button of the first application, it can be understood as user downloads first application, wherein the mark of the first application
Symbol can refer to Apply Names, such as " 360 mobile phone assistant ", " 360 mobile guards ", be also not limited to this certainly.
In an embodiment of the present invention, the second application refers to the application for having recommendation relation with the first application, for example with
Existing proposed algorithm can precompute and multiple first applications of second application with recommendation relation, wherein proposed algorithm
Refer to that existing commending system can be according to different dimensions（Such as content-based recommendation, the recommendation based on collaborative filtering, base
Recommend in correlation rule etc.）, applied according to the second application for the one or more first that user recommended user may like.
Below by taking the proposed algorithm based on collaborative filtering as an example, how description is established between the first application and the second application
Recommendation relation：
Specifically, the recommended flowsheet of collaborative filtering includes：
Step 1, generation user's rating matrix：In the commending system based on collaborative filtering, the score data of user can be with
It is expressed as a m * n matrix, m rows represent m user, and n row represent n resource item（Equivalent to application）, matrix element Ri, j
Represent user UiTo project IjScoring.Scoring can be expressed as 0-1 systems, and 0 represents not liking or do not buy（Equivalent to not pacifying
Dress）；1 represents to like or bought（Equivalent to having installed）；5 points of systems can also be expressed as, are represented according to preference from 0 to 5.
Step 2, selection method for measuring similarity, calculate user's scoring vector similitude.Similarity measures formula has as follows
Four kinds：
1）Cosine similarity metric formula（Cosine-based Simlarity）
2）Modified cosine similarity metric formula（Adjusted Cosine-based Simlarity）, i.e. Spear phases
Relation number
3）Associated similarity measure formulas（Correlation-based Simlarity）, i.e. Pearson correlation coefficient
4）Jaccard related coefficient measure formulas
Step 3, obtain similar application.The definite of similar application mainly has two methods：Threshold method and Top-N arest neighbors
Method.Threshold method is to set a threshold value for similarity, and the application that similarity is more than the threshold value is used as similar application.Top-N
Arest neighbors method is by the descending arrangement of similarity, chooses application of the similarity in top n as answering with recommendation relation
With.
Alternatively, the collaborative filtering based on article can be used in an embodiment of the present invention（ITEM-BASED
CF）, establish the recommendation relation between the first application and the second application.
For example, the application data of multiple users, including the data of the installation situation of application are input in recommended engine, should
Recommended engine can use the collaborative filtering based on article（ITEM-BASED CF）Computing is carried out, is then drawn by the recommendation
Hold up output application-apply recommendation list.
In order to improve the arithmetic speed of recommended engine, can answer in an embodiment of the present invention being input to recommended engine
Pre-processed with data, wherein pretreatment includes：Filtration treatment and IDization processing, wherein
Filtration treatment refers to：Duplicate customer or the insignificant user of filtering are filtered, wherein, filtering duplicate customer refers to：Sentence
It is disconnected whether to have duplicate customer, if so, then carrying out filtration treatment to duplicate customer；Insignificant user is filtered to refer to：Obtain user
Grade, if the grade of user is less than grade threshold, to the user carry out filtration treatment.Can be effective by filtration treatment
Data volume is reduced, avoids repetitive operation.
IDization processing refers to：Scaling method application data is become using existing character string and optimizes processing, such as MD5 is calculated
Method, is unique ID by more complicated data transformation, to reduce the operand of recommended engine.
Such as：Include using data：
IMEI bags name 1 | version number, bag name 2 | version number MID
68a25668a774a67ecf2b32d581b83525com.tencent.qqgame.qqlordhvga|20110,
com.sec.android.app.phoneutil|1,com.sec.a ndroid.app.kieswifi|1,,,,
98016185542f69c0c0878d501f1ac5b5
Data after IDization processing：
Soft_id imei：112268a25668a774a67ecf2b32d581b83525
It is as shown in table 1 below, have recorded the situation of five users application installation, wherein user by respective IMEI come area
Point；" 1 " represents that this has been installed using user, and " 0 " represents that this is not installed using user.
As shown in table 1, the user installation of " IMEI1 " QQ, bodyguard, assistant；The user installation of " IMEI2 " QQ, bodyguard,
Assistant, whirlwind run, footpath between fields footpath between fields；The user installation of " IMEI3 " QQ, assistant, Netease's news；The user installation of " IMEI4 " helps
Hand, Netease's news；The user installation of " IMEI5 " QQ, bodyguard.
| IMEI | Bodyguard | Assistant | Whirlwind is run | Footpath between fields footpath between fields | Netease's news | |
| 1 | 1 | 1 | 1 | 0 | 0 | 0 |
| 2 | 1 | 1 | 1 | 1 | 1 | 0 |
| 3 | 1 | 0 | 1 | 0 | 0 | 1 |
| 4 | 0 | 0 | 1 | 0 | 0 | 1 |
| 5 | 1 | 1 | 0 | 0 | 0 | 0 |
Of course, it should be understood that the situation of six kinds of applications of each self installation of five users recorded in table 1, selected six
Kind application includes：QQ, bodyguard, whirlwind run race, footpath between fields footpath between fields and Netease's news, are also not limited to this certainly.Usually, in order to improve
The accuracy of collaborative filtering result based on article, can further extend the field of the investigation.
The situation for having installed application of five users according to upper table 1, using similarity measurements quantity algorithm, can calculate
Go out the similarity between two applications.Alternatively, in an embodiment of the present invention, can be calculated using equation below between applying
Similarity Simlarity
Wherein, x is represented：Each user installation application X（Such as QQ）Situation, " 1 " represent installed, " 0 " represent do not pacify
Dress；
Y is represented：Each user installation application Y（Such as bodyguard）Situation, " 1 " represent installed, " 0 " represent do not install；
Below exemplified by calculating QQ and bodyguard's similarity,
As shown in upper table 1, the installation situation of five user QQ is understood, x=（1,1,1,0,1）
As shown in upper table 1, the installation situation of five user bodyguards is understood, y=（1,1,0,0,1）, with reference to above-mentioned similarity
Calculation formula can obtain：
Simlarity（X, y）=3/SQRT（4×3）=0.866025
That is, the similarity of QQ and bodyguard are 0.866025, and the calculating process of the similarity between other application is similar,
No longer apply and state herein.
It is as shown in table 2 below, to combine table 1 and similarity measurements quantity algorithm, the application similarity being calculated
| APP | Bodyguard | Assistant | Whirlwind is run | Footpath between fields footpath between fields | Netease's news | |
| 1.000000 | 0.866025 | 0.750000 | 0.500000 | 0.500000 | 0.353553 | |
| Bodyguard | 0.866025 | 1.000000 | 0.577350 | 0.577350 | 0.577350 | 0.000000 |
| Assistant | 0.750000 | 0.577350 | 1.000000 | 0.500000 | 0.500000 | 0.707106 |
| Whirlwind is run | 0.500000 | 0.577350 | 0.500000 | 1.000000 | 1.000000 | 0.000000 |
| Footpath between fields footpath between fields | 0.500000 | 0.577350 | 0.500000 | 1.000000 | 1.000000 | 0.000000 |
| Netease's news | 0.353553 | 0.000000 | 0.707106 | 0.000000 | 0.000000 | 1.000000 |
In an embodiment of the present invention, the similarity between two applications is higher, illustrates to be mounted with one of application
User, installs the possibility bigger of another application, therefore two applications can be established based on the similarity between two applications
Between recommendation relation.
Based on the similarity between each application shown in above-mentioned table 2, sort, obtain such as following table from high to low according to similarity
The table for respectively applying similarity shown in 3
For example, for QQ, similarity is included by the application of height：Bodyguard, assistant, whirlwind run race, footpath between fields footpath between fields and Netease
News.
For bodyguard, similarity is included by the application of height：QQ, assistant, whirlwind run race, footpath between fields footpath between fields and Netease's news.
In an embodiment of the present invention, the definite of recommendation relation application can also use：Threshold method and Top-N arest neighbors
Method.Wherein
Threshold method is to set a threshold value for similarity, and application of the similarity more than the threshold value, which is used as, has recommendation relation
Application.Such as threshold value is arranged to 0.7, application of the similarity more than 0.7 is as the application with recommendation relation.Come for QQ
Say, the application with recommendation relation includes：Bodyguard and assistant.For assistant, the application with recommendation relation includes：QQ and
Netease's news.
Top-N arest neighbors methods are by the descending arrangement of similarity, and choosing similarity and being used as in the application of top n has
The application of recommendation relation.Such as N set 3, similarity first 3 application as have recommendation relation application, for QQ
For, the application with recommendation relation includes：Bodyguard, assistant and whirlwind are run.
Alternatively, before step S101, network side server can browse the first application in response to user and download the
The behavior of one application, records the identifier of the first application browsed and record are downloaded first in the daily record of network side server
The identifier of application.
Alternatively, terminal equipment side can be by the behavior of the behavior for browsing the first application of user and the application of download first
Network side server is reported, first downloaded by the identifier and record of the browse first application of network side server record user
The identifier of application, to carry out user behavior analysis.
Specifically, the identifier for the first application that user is currently browsing is recorded in the daily record of network side server,
And the identifier with second application of first application with recommendation relation.User is being recorded in the daily record of network side server just
In the identifier of the first application of download, and the identifier with second application of first application with recommendation relation.
Alternatively, the form of the daily record can be the journal format of Hadoop, and daily record is a line one, and journal format can be with
It is described as successively：Date, time, rank, associated class and prompt message etc..
Correspondingly, before step S101, the daily record in network side server is analyzed one by one；Then judge daily record whether be
Recommend the daily record of application；If recommending the daily record of application, then into the mark for the first application that acquisition user is browsed and downloads
Know symbol, and there is the step of identifier that the second of recommendation relation applies with the described first application.
Step S103, the number of visits and download time of the application of statistics first.
Alternatively, in an embodiment of the present invention, the page access amount (Page of the page where the first application can be counted
View), to realize the number of visits of the application of statistics first, this is also not limited to certainly.
Alternatively, in an embodiment of the present invention, the download button of the first application can be counted by point by counter
The number hit, to realize the download time of the application of statistics first, is also not limited to this certainly.
Alternatively, after step s 103, this method 100 further includes：According to pushing away between the first application and the second application
Recommend relation, preserve the correspondence of the identifier and the second identifier applied of the first application, and it is browsed or downloaded
The number of visits and download time of first application.
Specifically, the number of visits of the first application and download time can be preserved according to following form：The mark of second application
Know symbol, the identifier of the first application, the number of visits of the first application and the download time of the first application.
Step S105, according to the first of statistics application download time and number of visits, calculate according to first application and
The recommendation relation of second application downloads the download conversion ratio of the first application.
In an embodiment of the present invention, first application download conversion ratio can be used for weigh user browsing the first application
Afterwards, the probability of first application can be downloaded, alternatively, the download conversion ratio of first application can be calculated using the following formula
Arrive：
Download conversion ratio=download time/number of visits
Wherein, the download conversion ratio of the first application is bigger, then illustrates that user after the application is browsed, downloads the general of the application
Rate is bigger.It is smaller to download conversion ratio, then illustrates user after the application is browsed, the probability for downloading the application is smaller.Therefore, usually
In the case of, it can will download the big application of conversion ratio and recommend user, the success rate downloaded with improving application to recommend.
In an embodiment of the present invention, mapreduce average hadoop can be passed through（It is a kind of programming model, is used for
The concurrent operation of large-scale dataset）Technology extracts the relevant information for analyzing user behavior from the daily record of server, when
So it is also not limited to this.In order to improve the accuracy of analysis, the daily record in predetermined time period can be analyzed, for example, it is right
The daily record of 10 days is analyzed and handled, and processing method is as follows：
Step 1, judge whether the daily record in network side server is the daily record for recommending application one by one, if this daily record is recommendation
The daily record of application, then obtain the identifier for browsing the first application in daily record or download the identifier of the first application, and obtains
With the identifier of second application of first application with recommendation relation.Wherein second application has recommendation relation with the first application,
Represent that the second application has higher similarity with the first application.
Step 2, the number of visits and download time of the application of statistics first.
Alternatively, the number of visits of the first application and download time can be preserved according to following form：The mark of second application
Know symbol, the identifier of the first application, the number of visits of the first application and the download time of the first application.
Step 3, number of visits and download time according to the first application, can obtain according to the first application and the second application
Recommendation relation download first application download conversion ratio.
In an embodiment of the present invention, first application download conversion ratio can be used for weigh user browsing the first application
When, download first probability applied.The download conversion ratio of first application can be calculated using the following formula：
Download conversion ratio=download time/number of visits
First application download conversion ratio it is bigger, then illustrate user browse this first in application, download this first application
Probability it is bigger.And first application download conversion ratio it is smaller, then illustrate user browse this first in application, download this first
The probability of application is smaller.Therefore, it is generally the case that can will download the first big application of conversion ratio and recommend user, to improve
The success rate downloaded is recommended in first application.
As shown in Fig. 2, to recommend the two of the flow chart of the method for application in the embodiment of the present invention based on user behavior, should
Method 200 includes：
Step S201, the identifier for the first application that user browses and the identifier of the first application of download are obtained, and
With the identifier of second application of first application with recommendation relation.
In an embodiment of the present invention, the first application and the second application refer to possess unique identifier（Such as digital sequence
Row number）Software, which can include the software suitable for computing device and the software suitable for mobile equipment.
In the embodiment of the present invention, the navigation patterns for browsing the first application that can be based on user and the application of download first
Download behavior, in network side server（Such as log server or recommendation server）The first application that middle record user browses
Identifier and download first application identifier.Wherein, user browses the application displaying of the first application on the terminal device
The situation of the page, it can be understood as user browses first application, this has the description of the first application using record on displayed page
Information, the description information include：The identifier of first application, the icon of the first application, the download address of the first application；For point
Hit the situation for the download button for downloading the first application, it can be understood as user downloads first application, wherein the mark of the first application
Apply Names, such as " 360 mobile phone assistant ", " 360 mobile guards " can be referred to by knowing symbol, be also not limited to this certainly.
In an embodiment of the present invention, the second application refers to the application for having recommendation relation with the first application, for example with
Existing proposed algorithm can precompute and multiple first applications of second application with recommendation relation, wherein proposed algorithm
Refer to that existing commending system can be according to different dimensions（Such as content-based recommendation, the recommendation based on collaborative filtering, base
Recommend in correlation rule etc.）, it is the application that user recommended user may like according to the second application.
Alternatively, the collaborative filtering based on article can be used in an embodiment of the present invention（ITEM-BASED
CF）Calculate and multiple first applications of second application with recommendation relation.
Alternatively, before step S201, network side server can browse the first application in response to user and download the
The behavior of one application, records the identifier of the first application browsed and record are downloaded first in the daily record of network side server
The identifier of application.
Alternatively, terminal equipment side can be by the behavior of the behavior for browsing the first application of user and the application of download first
Network side server is reported, first downloaded by the identifier for browsing the first application and record of network side server record user should
Identifier, to carry out user behavior analysis.
Specifically, the identifier for the first application that user is currently browsing is recorded in the daily record of network side server,
And the identifier with second application of first application with recommendation relation.User is being recorded in the daily record of network side server just
In the identifier of the first application of download, and the identifier with second application of first application with recommendation relation.
Alternatively, the form of the daily record can be the journal format of Hadoop, and daily record is a line one, and journal format can be with
It is described as successively:Date, time, rank, associated class and prompt message etc..
Correspondingly, before step S201, the daily record in network side server is analyzed one by one；Then judge daily record whether be
Recommend the daily record of application；If recommending the daily record of application, then into the mark for the first application that acquisition user is browsed and downloads
Know symbol, and there is the step of identifier that the second of recommendation relation applies with the described first application.
Step S203, the number of visits and download time of the application of statistics first.
Alternatively, in an embodiment of the present invention, the page access amount (Page of the page where the first application can be counted
View), to realize the number of visits of the application of statistics first, this is also not limited to certainly.
Alternatively, in an embodiment of the present invention, the download button of the first application can be counted by point by counter
The number hit, to realize the download time of the application of statistics first, is also not limited to this certainly.
Alternatively, after step S203, this method 200 further includes：According to pushing away between the first application and the second application
Recommend relation, preserve the correspondence of the identifier and the second identifier applied of the first application, and it is browsed or downloaded
The number of visits and download time of first application.
Specifically, the number of visits of the first application and download time can be preserved according to following form：The mark of second application
Know symbol, the identifier of the first application, the number of visits of the first application and the download time of the first application.
Step S205, according to the first of statistics application download time and number of visits, calculate according to first application and
The recommendation relation of second application downloads the download conversion ratio of the first application.
In an embodiment of the present invention, first application download conversion ratio can be used for weigh user browsing the first application
Afterwards, the probability of first application is downloaded, the download conversion ratio of first application can be calculated using the following formula：
Download conversion ratio=download time/number of visits
Wherein, first application download conversion ratio it is bigger, then illustrate user browse this first application after, download this first
The probability of application is bigger.And the download conversion ratio of the first application is smaller, then illustrate user after first application is browsed, downloading should
The probability of first application is smaller.Therefore, it is generally the case that it can will download the first big application of conversion ratio and recommend user, with
Improve the success rate that the first application is recommended to download.
Step S207, the download of receiving terminal apparatus side or the request message of the second application is browsed, request message includes
The identifier of second application.
Step S209, the request of second application is browsed or downloaded in response to user, and in recommendation server side, generation should
With the page is recommended, the application displayed page includes：Using region and application recommendation region is downloaded, wherein application is downloaded in region
Including：The description information of second application, includes using region is recommended：With one or more the of the second application with recommendation relation
The description information of one application, and the download conversion ratio of each first application.
Alternatively, in an embodiment of the present invention, the description information of the second application includes：The identifier of second application, the
The icon of two applications, second download address applied etc..
Alternatively, in an embodiment of the present invention, the description information of the first application includes：The identifier of first application, the
The icon of one application, the download address of the first application, the number of visits of the first application, first download time applied etc..
Further, the description information of first application further includes：The rationale for the recommendation of first application.Alternatively, the recommendation
Reason can be filled in by user, and be stored in network side server.
In a kind of mode of the embodiment of the present invention, the description that the multiple first applications are included on application displayed page is believed
Breath, the description information of multiple first applications are arranged in order according to the size of the download conversion ratio of the first application.
In the another way of the embodiment of the present invention, one first mark applied is included on application displayed page
Symbol, this one first download conversion ratio applied is maximum, it is achieved thereby that recommending unique one first to answer based on the second application
With so that user considers whether to download.
Step S211, terminal device will be sent to using displayed page in recommendation application server side, and in terminal device
Upper display.
In an embodiment of the present invention, user can apply displayed page by what is shown on terminal device, understand and recommend
Which Downloadable first application server recommends to the user, and the recommendation region on application displayed page further includes
The download conversion ratio of first application, by the download conversion ratio of the first application, allows active user to understand other users and is browsing
After first application, the probability of first application can be downloaded.Such as in the multiple first applications, if the download of the first application A
Conversion ratio is maximum, can reflect that this first has preferable user experience using A, because user applies A having browsed first
Afterwards, first can be downloaded and apply A, therefore user's selection from the multiple first applications can be helped to download the by downloading conversion ratio
One applies A, and then improves the download success rate of the first application A.
Above-mentioned flow is introduced below by specific mode：
Step A, the download application based on all proposed algorithms and the download row of browse application are taken out in daily daily record
For and navigation patterns.Based on step A, the download conversion ratio that the first application of the first application is recommended in the second application can be obtained.Can
Referring to step S101~step S105.
Step B, nearly predetermined time period is selected（Such as 10 days）Download conversion ratio, weighted sum obtains the predetermined time
The download conversion ratio of first application of section, by this first download conversion ratio applied, obtains in predetermined time period, and second should
Download conversion ratio with which specific application in multiple first applications of recommendation is higher.
Step C, when recommending multiple first application associated with the second application, lower idling in statistics is just selected
Rate it is high first application come before, download conversion ratio it is low first application come behind.
Step D, when carrying out first using unique recommend, can be answered downloading multiple first of conversion ratio more than 0
First application to be recommended is selected as weight according to download conversion ratio by the use of middle.
As shown in figure 3, for a kind of recommendation server block diagram in the embodiment of the present invention, which includes：
Acquisition module 301, the identifier of the first application browsed for obtaining user and the mark for the first application downloaded
Symbol, and the identifier with second application of described first application with recommendation relation；Alternatively, first application and second
Recommendation relation between is generated by the collaborative filtering based on article.
Statistical module 303, for counting the number of visits and download time of the first application；
Conversion ratio generation module 305 is downloaded, for the download time and number of visits of first application according to statistics,
Calculate the download conversion ratio that first application is downloaded according to the recommendation relation of the second application and the first application.
Alternatively, in another embodiment of the present invention, the recommendation server 300 further includes：Preserving module, is used for
Recommendation relation between being applied according to the described first application and second, what the identifier and second that preservation described first is applied were applied
The correspondence of identifier, and the first number of visits applied and download time each browsed or downloaded.
Alternatively, in another embodiment of the present invention, the recommendation server 300 further includes：Log analysis module,
For judge the daily record whether be recommend application daily record；If recommending the daily record of application, then acquisition module is triggered.
Alternatively, in another embodiment of the present invention, the recommendation server 300 further includes：Browse logging modle,
For browsing the behavior of the first application in response to user, recommend application daily record in record user currently browsing first
The identifier of application, and the identifier with second application of described first application with recommendation relation.
Alternatively, in another embodiment of the present invention, the recommendation server 300 further includes：Download History module,
For downloading the behavior of first application, record user is downloading in the daily record for recommending application first in response to user
The identifier of application, and the identifier with second application of described first application with recommendation relation.
Alternatively, in another embodiment of the present invention, the recommendation server 300 further includes：
Using page generation module 307 is recommended, for the request of second application, life to be browsed or downloaded in response to user
Recommend the page into application, the application displayed page includes：Using region and application recommendation region is downloaded, wherein using download area
Domain includes：The description information of second application, includes using region is recommended：With one or more of the second application with recommendation relation
The description information of a first application, and the download conversion ratio of each first application；
Using page sending module 309 is recommended, for the application displayed page to be sent to terminal device, and described
Shown on terminal device.
Alternatively, it is described using on displayed page include it is multiple it is described first application identifiers, the multiple first
The identifier of application is arranged in order according to the size of the download conversion ratio of the described first application.
Alternatively, described using the identifier for including one first application on displayed page, one first application
Download conversion ratio it is maximum.
Alternatively, in an embodiment of the present invention, the description information of first application includes：The mark of first application
Symbol, the icon of the first application, the download address of the first application, the number of visits of the first application and the download time of the first application；
The description information of second application includes：The identifier of second application, the icon of the second application and the second application
Download address.
Alternatively, in an embodiment of the present invention, the description information of first application further includes：The recommendation of first application
Reason.
A kind of system for recommending application based on user behavior is additionally provided in another aspect of this invention, as shown in figure 4, should
System includes recommendation server 300 as described above, and at least one terminal device 43 being connected with recommendation server 300.
As shown in figure 5, to recommend the block schematic illustration of the system of application in the embodiment of the present invention based on user behavior, should
System is broadly divided into five-layer structure：（1）Product represent layer：What user saw inside mobile phone assistant applies displayed page；（2）Clothes
Business end front end interface：For the data that interface returns will to be recommended to Reseal into the form that front end can be shown；（3）Recommending data connects
Mouthful：According to different algorithms, different algorithms is recommended data；（4）Recommended engine：According to different algorithms, recommend data；
（4）Data Layer：The initial data of application, including Apply Names, using label, application class etc.；User's gets data ready, including
The behavior that behavior that user browses, user download.
Algorithm and display be not inherently related to any certain computer, virtual system or miscellaneous equipment provided herein.
Various general-purpose systems can also be used together with teaching based on this.As described above, required by constructing this kind of system
Structure be obvious.In addition, the present invention is not also directed to any certain programmed language.It should be understood that it can utilize various
Programming language realizes the content of invention described herein, and the description done above to language-specific is to disclose this hair
Bright preferred forms.
In the specification that this place provides, numerous specific details are set forth.It is to be appreciated, however, that the implementation of the present invention
Example can be put into practice in the case of these no details.In some instances, known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this description.
Similarly, it will be appreciated that in order to simplify the disclosure and help to understand one or more of each inventive aspect,
Above in the description to the exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the method for the disclosure should be construed to reflect following intention：I.e. required guarantor
The application claims of shield features more more than the feature being expressly recited in each claim.It is more precisely, such as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following embodiment are expressly incorporated in the embodiment, wherein each claim is in itself
Separate embodiments all as the present invention.
Those skilled in the art, which are appreciated that, to carry out adaptively the module in the equipment in embodiment
Change and they are arranged in one or more equipment different from the embodiment.Can be the module or list in embodiment
Member or component be combined into a module or unit or component, and can be divided into addition multiple submodule or subelement or
Sub-component.In addition at least some in such feature and/or process or unit exclude each other, it can use any
Combination is to this specification（Including adjoint claim, summary and attached drawing）Disclosed in all features and so disclosed appoint
Where all processes or unit of method or equipment are combined.Unless expressly stated otherwise, this specification（Including adjoint power
Profit requirement, summary and attached drawing）Disclosed in each feature can be by providing identical, equivalent or similar purpose alternative features come generation
Replace.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included some features rather than further feature, but the combination of the feature of different embodiments means in of the invention
Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed
One of meaning mode can use in any combination.
The all parts embodiment of the present invention can be with hardware realization, or to be run on one or more processor
Software module realize, or realized with combinations thereof.It will be understood by those of skill in the art that it can use in practice
Microprocessor or digital signal processor（DSP）Come realize it is according to embodiments of the present invention based on user behavior recommend application
The some or all functions of some or all components in system.The present invention is also implemented as being used to perform being retouched here
The some or all equipment or program of device for the method stated（For example, computer program and computer program product）.
Such program for realizing the present invention can store on a computer-readable medium, or can have one or more signal
Form.Such signal can be downloaded from internet website and obtained, either provide on carrier signal or with it is any its
He provides form.
It should be noted that the present invention will be described rather than limits the invention for above-described embodiment, and ability
Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between bracket should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" before element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of some different elements and being come by means of properly programmed computer real
It is existing.In the unit claim of some equipment is listed, several in these equipment can be by same hardware branch
To embody.The use of word first, second, and third does not indicate that any order.These words can be explained and run after fame
Claim.
The invention also discloses A1, a kind of method for recommending application based on user behavior, including：
The identifier of the first application for obtaining the identifier for the first application that user browses and downloading, and with described first
Using the identifier with the second application for recommending relation；
Count the number of visits and download time of first application；
According to the download time and number of visits of first application of statistics, calculating should according to the second application and first
Relation is recommended to download the download conversion ratio of first application.
A2, the method according to A1, before download conversion ratio is calculated, the method further includes：
According to the recommendation relation between the described first application and the second application, the identifier and the of first application is preserved
The correspondence of the identifier of two applications, and the number of visits and download time of the first application.
A3, the method according to A1, wherein, the recommendation relation between first application and the second application is by based on thing
The collaborative filtering generation of product.
A4, the method according to A1, wherein, the method further includes：
The behavior of the first application is browsed in response to user, user is recorded in the daily record of the recommendation application of network side server
The identifier of current the first application browsed, and the mark with second application of described first application with recommendation relation
Symbol.
A5, the method according to A4, wherein, the method further includes：
The behavior of first application is downloaded in response to user, in the daily record of the recommendation application of the network side server
The identifier for the first application that record user is downloading, and applied with second of the described first application with recommendation relation
Identifier.
A6, the method according to A1, wherein, the identifier for obtaining the first application that user browses and download the
One application identifier, and with described first application with recommendation relation second application identifier the step of before, institute
The method of stating further includes：
Judge whether the daily record in the network side server is the daily record for recommending application one by one；
If recommending the daily record of application, then enter and obtain the browsed identifier with the downloaded first application of user, with
And there is the step of identifier of the second application of recommendation relation with the described first application.
A7, the method according to A1, the method further include：
The request of second application is browsed or downloaded in response to user, and generating application in recommendation server side recommends page
Face, the application displayed page include：Using region and application recommendation region is downloaded, include wherein region is downloaded in application：The
The description information of two applications, includes using region is recommended：Applied with the second one or more first of the application with recommendation relation
Description information, and it is each first application download conversion ratio；
The application displayed page is sent to terminal device in the recommendation application server side, and is set in the terminal
Standby upper display.
A8, the method according to A7, wherein, when described using the description information for having multiple first applications on displayed page
When, the description information of the multiple first application is arranged in order according to the size of the download conversion ratio of the described first application.
A9, the method according to A7, wherein, the description information of first application includes：The identifier of first application,
The icon of first application, the download address of the first application, the number of visits of the first application and the download time of the first application；
The description information of second application includes：The identifier of second application, the icon of the second application and the second application
Download address.
A10, the method according to A9, wherein, the description information of first application further includes：The recommendation of first application
Reason.
B11, a kind of recommendation server, including：
Acquisition module, the identifier of the first application browsed for obtaining user and the identifier for the first application downloaded,
And the identifier with second application of described first application with recommendation relation；
Statistical module, for counting the number of visits and download time of first application；
Conversion ratio generation module is downloaded, for the download time and number of visits of first application according to statistics, meter
Calculate the download conversion ratio that first application is downloaded according to the recommendation relation of the second application and the first application.
B12, the recommendation server according to B11, the recommendation server further include：
Preserving module, for according to the recommendation relation between the described first application and the second application, preserving described first should
The correspondence of identifier and the identifier of the second application, and the number of visits and download time of the first application.
B13, the recommendation server according to B11, wherein, the recommendation relation between first application and the second application
Generated by the collaborative filtering based on article.
B14, the recommendation server according to B11, the recommendation server further include：
Log analysis module, for judge the daily record whether be recommend application daily record；If recommending the daily record of application,
Then trigger the acquisition module.
B15, the recommendation server according to B14, wherein, the recommendation server further includes：
Logging modle is browsed, for browsing the behavior of the first application in response to user, is recorded in the daily record for recommending application
The identifier for the first application that user is currently browsing, and applied with second of the described first application with recommendation relation
Identifier.
B16, the recommendation server according to B14, wherein, the recommendation server further includes：
Download History module, for downloading the behavior of first application in response to user, in the daily record for recommending application
The identifier for the first application that record user is downloading, and applied with second of the described first application with recommendation relation
Identifier.
B17, the recommendation server according to B11, the recommendation server further include：
Using page generation module is recommended, for the request of second application, generation to be browsed or downloaded in response to user
Using the page is recommended, the application displayed page includes：Using region and application recommendation region is downloaded, wherein region is downloaded in application
Include：The description information of second application, includes using region is recommended：With the second one or more of the application with recommendation relation
The description information of first application, and the download conversion ratio of each first application；
Using page sending module is recommended, for the application displayed page to be sent to terminal device, and at the end
Shown in end equipment.
B18, the recommendation server according to B17, wherein, include multiple described first on the application displayed page
The identifier of application, the identifier of the multiple first application are arranged successively according to the size of the download conversion ratio of the described first application
Row.
B19, the recommendation server according to B17, wherein, the description information of first application includes：First application
Identifier, first application icon, first application download address, first application number of visits and first application download
Number；
The description information of second application includes：The identifier of second application, the icon of the second application and the second application
Download address.
B20, the recommendation server according to claim B19, wherein, the description information of first application is also wrapped
Include：The rationale for the recommendation of first application.
C21, a kind of system for recommending application based on user behavior, including the recommendation service as described in B11~B20 is any
Device.
Claims (21)
1. a kind of method for recommending application based on user behavior, including：
The identifier for the first application that acquisition user browses and the identifier for the first application downloaded, and applied with described first
Identifier with the second application for recommending relation；
Count the number of visits and download time of first application；
According to the download time and number of visits of first application of statistics, calculate what is applied according to the second application and first
Recommendation relation downloads the download conversion ratio of first application.
2. according to the method described in claim 1, before download conversion ratio is calculated, the method further includes：
According to the recommendation relation between the described first application and the second application, the identifier and second for preserving first application should
The correspondence of identifier, and the number of visits and download time of the first application.
3. according to the method described in claim 1, wherein, recommendation relation between first application and the second application by based on
The collaborative filtering generation of article.
4. according to the method described in claim 1, wherein, the method further includes：
The behavior of the first application is browsed in response to user, it is current to record user in the daily record of the recommendation application of network side server
The identifier of the first application browsed, and the identifier with second application of described first application with recommendation relation.
5. according to the method described in claim 4, wherein, the method further includes：
The behavior of first application is downloaded in response to user, is recorded in the daily record of the recommendation application of the network side server
The identifier for the first application that user is downloading, and the mark with second application of described first application with recommendation relation
Symbol.
6. according to the method described in claim 1, wherein, the identifier for obtaining the first application that user browses and download
First application identifier, and with described first application with recommendation relation second application identifier the step of before,
The method further includes：
Judge whether the daily record in network side server is the daily record for recommending application one by one；
If recommending the daily record of application, then into the identifier for the first application that acquisition user is browsed and downloads, Yi Jiyu
The step of identifier of the second application of first application with recommendation relation.
7. according to the method described in claim 1, the method further includes：
The request of second application is browsed or downloaded in response to user, and in recommendation server side, displayed page, institute are applied in generation
State includes using displayed page：Using region and application recommendation region is downloaded, include wherein region is downloaded in application：Second application
Description information, using recommend region include：The description applied with the second one or more first of the application with recommendation relation
Information, and the download conversion ratio of each first application；
The application displayed page is sent to terminal device in the recommendation server side, and is shown on the terminal device
Show.
8. according to the method described in claim 7, wherein, believe when described using the description for there are multiple first applications on displayed page
During breath, the description information of the multiple first application is arranged in order according to the size of the download conversion ratio of the described first application.
9. according to the method described in claim 7, wherein, the description information of first application includes：The mark of first application
Symbol, the icon of the first application, the download address of the first application, the number of visits of the first application and the download time of the first application；
The description information of second application includes：Under the identifier of second application, the icon of the second application and the second application
Set address.
10. according to the method described in claim 9, wherein, the description information of first application further includes：First application pushes away
Recommend reason.
11. a kind of recommendation server, including：
Acquisition module, the identifier of the first application browsed for obtaining user and the identifier for the first application downloaded, and
With the identifier of second application of described first application with recommendation relation；
Statistical module, for counting the number of visits and download time of first application；
Conversion ratio generation module is downloaded, for the download time and number of visits of first application according to statistics, is calculated
The download conversion ratio of first application is downloaded according to the recommendation relation of the second application and the first application.
12. recommendation server according to claim 11, the recommendation server further includes：
Preserving module, for according to the recommendation relation between the described first application and the second application, preserving first application
The correspondence of identifier and the identifier of the second application, and the number of visits and download time of the first application.
13. recommendation server according to claim 11, wherein, the recommendation between first application and the second application is closed
System is generated by the collaborative filtering based on article.
14. recommendation server according to claim 11, the recommendation server further includes：
Log analysis module, for judge the daily record in network side server whether be recommend application daily record；Should if recommending
Daily record, then trigger the acquisition module.
15. recommendation server according to claim 14, wherein, the recommendation server further includes：
Logging modle is browsed, for browsing the behavior of the first application in response to user, user is recorded in the daily record for recommending application
The identifier of current the first application browsed, and the mark with second application of described first application with recommendation relation
Symbol.
16. recommendation server according to claim 14, wherein, the recommendation server further includes：
Download History module, for downloading the behavior of first application in response to user, records in the daily record for recommending application
The identifier for the first application that user is downloading, and the mark with second application of described first application with recommendation relation
Symbol.
17. recommendation server according to claim 11, the recommendation server further includes：
Using displayed page generation module, for the request of second application, generation application to be browsed or downloaded in response to user
Displayed page, the application displayed page include：Using region and application recommendation region is downloaded, wherein region Zhong Bao is downloaded in application
Include：The description information of second application, includes using region is recommended：With the second one or more first of the application with recommendation relation
The description information of application, and the download conversion ratio of each first application；
Using displayed page sending module, for the application displayed page to be sent to terminal device, and set in the terminal
Standby upper display.
18. recommendation server according to claim 17, wherein, multiple described the are included on the application displayed page
One application identifier, it is the multiple first application identifier according to described first application download conversion ratio size successively
Arrangement.
19. recommendation server according to claim 17, wherein, the description information of first application includes：First should
Under identifier, the icon of the first application, the download address of the first application, the number of visits of the first application and the first application
Carry number；
The description information of second application includes：Under the identifier of second application, the icon of the second application and the second application
Set address.
20. recommendation server according to claim 19, wherein, the description information of first application further includes：First
The rationale for the recommendation of application.
21. a kind of system for recommending application based on user behavior, including the recommendation service as described in claim 11~20 is any
Device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410014009.8A CN103763361B (en) | 2014-01-13 | 2014-01-13 | A kind of method, system and recommendation server for recommending application based on user behavior |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410014009.8A CN103763361B (en) | 2014-01-13 | 2014-01-13 | A kind of method, system and recommendation server for recommending application based on user behavior |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103763361A CN103763361A (en) | 2014-04-30 |
| CN103763361B true CN103763361B (en) | 2018-04-27 |
Family
ID=50530520
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410014009.8A Active CN103763361B (en) | 2014-01-13 | 2014-01-13 | A kind of method, system and recommendation server for recommending application based on user behavior |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103763361B (en) |
Families Citing this family (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104158865A (en) * | 2014-08-01 | 2014-11-19 | 北京奇虎科技有限公司 | Method and system for pushing applications to terminal |
| CN104239421B (en) * | 2014-08-21 | 2018-01-16 | 北京奇虎科技有限公司 | A kind of method and system for pushing VAS application -to-terminal service |
| CN104270429B (en) * | 2014-09-19 | 2017-12-08 | 北京奇虎科技有限公司 | A kind of method and apparatus for pushing VAS application -to-terminal service |
| US20160132780A1 (en) * | 2014-11-12 | 2016-05-12 | Google Inc. | Automatic Selection of Images for an Application |
| CN104572962A (en) * | 2014-12-31 | 2015-04-29 | 浙江大学 | APP (Application) recommendation method and system |
| CN104537104A (en) * | 2015-01-13 | 2015-04-22 | 百度在线网络技术（北京）有限公司 | Software ranking method and system |
| CN104935651A (en) * | 2015-06-03 | 2015-09-23 | 魅族科技（中国）有限公司 | Content recommending method and terminal |
| CN104994147B (en) * | 2015-06-24 | 2018-04-17 | 上海卓悠网络科技有限公司 | The recommendation method of software download, system and the server-side being applicable in and user equipment |
| CN106452808A (en) * | 2015-08-04 | 2017-02-22 | 北京奇虎科技有限公司 | Data processing method and data processing device |
| CN105224614A (en) * | 2015-09-17 | 2016-01-06 | 北京金山安全软件有限公司 | Application program classification display method and device |
| CN105183882A (en) * | 2015-09-23 | 2015-12-23 | 百度在线网络技术（北京）有限公司 | Application software recommending method and device |
| CN105554799A (en) * | 2015-12-10 | 2016-05-04 | 小米科技有限责任公司 | Statistical data display method and apparatus |
| CN106210025B (en) * | 2016-07-05 | 2019-08-27 | 青岛海信传媒网络技术有限公司 | Application program download determines method and device |
| CN106161627B (en) * | 2016-07-12 | 2019-10-15 | 百度在线网络技术（北京）有限公司 | Method and device for pushed information |
| CN107807944A (en) * | 2016-09-09 | 2018-03-16 | 北京搜狗科技发展有限公司 | A kind of file recommendation method, device and electronic equipment |
| CN108268545B (en) * | 2016-12-30 | 2020-11-10 | 中移（杭州）信息技术有限公司 | Method and device for establishing hierarchical user label library |
| CN108268519B (en) * | 2016-12-30 | 2022-05-24 | 阿里巴巴集团控股有限公司 | Method and device for recommending network object |
| CN106815079B (en) * | 2016-12-30 | 2020-02-07 | Oppo广东移动通信有限公司 | Hot application control method and mobile terminal |
| CN106940710B (en) * | 2017-02-27 | 2020-03-24 | Oppo广东移动通信有限公司 | Information pushing method and device |
| CN110555155B (en) * | 2017-08-30 | 2023-04-07 | 腾讯科技（北京）有限公司 | Article information recommendation method, device and storage medium |
| CN107885852B (en) * | 2017-11-13 | 2021-09-10 | 上海交通大学 | APP recommendation method and system based on APP usage record |
| CN107967330A (en) * | 2017-11-27 | 2018-04-27 | 中山大学 | Method is recommended based on the mobile application for downloading behavioral data and vectorial representative learning |
| CN108021642B (en) * | 2017-11-29 | 2021-02-02 | Oppo广东移动通信有限公司 | Application program recommendation method and device, mobile terminal and storage medium |
| CN108038194B (en) * | 2017-12-11 | 2020-08-28 | Oppo广东移动通信有限公司 | Classification card generation method, system, server and computer readable storage medium |
| CN111666485B (en) * | 2019-03-07 | 2024-01-09 | 深圳市雅阅科技有限公司 | Information recommendation method, device and terminal |
| CN111586126B (en) * | 2020-04-28 | 2023-06-02 | 百度在线网络技术（北京）有限公司 | Applet pre-downloading method, device, equipment and storage medium |
| CN112182416B (en) * | 2020-09-10 | 2024-02-02 | 北京市科学技术情报研究所 | Scientific and technological resource dynamic collaborative filtering recommendation method based on user group preference |
| CN112131476B (en) * | 2020-09-27 | 2023-12-08 | 深圳市锐尔觅移动通信有限公司 | Application recommendation method, device, apparatus, terminal and readable storage medium |
| CN116861066A (en) * | 2022-03-25 | 2023-10-10 | 华为技术有限公司 | Application recommendation method and electronic equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722524A (en) * | 2012-05-07 | 2012-10-10 | 北京邮电大学 | Website recommendation result displaying method and device and terminal with the device |
| WO2012138574A1 (en) * | 2011-04-04 | 2012-10-11 | Qualcomm Incorporated | Recommending mobile content by matching similar users |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101431485B (en) * | 2008-12-31 | 2012-07-04 | 深圳市迅雷网络技术有限公司 | Method and system for automatically recommending internet information |
| CN102567352B (en) * | 2010-12-20 | 2015-04-29 | 联想(北京)有限公司 | Prompt method for application provided by websites and terminal equipment utilizing same |
| CN103049452B (en) * | 2011-10-14 | 2017-12-26 | 百度在线网络技术（北京）有限公司 | It is a kind of based on estimate download rate carry out using sequence method and apparatus |
| CN103020845B (en) * | 2012-12-14 | 2018-08-10 | 百度在线网络技术（北京）有限公司 | A kind of method for pushing and system of mobile application |
-
2014
- 2014-01-13 CN CN201410014009.8A patent/CN103763361B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012138574A1 (en) * | 2011-04-04 | 2012-10-11 | Qualcomm Incorporated | Recommending mobile content by matching similar users |
| CN102722524A (en) * | 2012-05-07 | 2012-10-10 | 北京邮电大学 | Website recommendation result displaying method and device and terminal with the device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103763361A (en) | 2014-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103763361B (en) | A kind of method, system and recommendation server for recommending application based on user behavior | |
| JP5767748B2 (en) | Tracking feeds on social networks | |
| US10162891B2 (en) | Determining demographics based on user interaction | |
| CN103530751B (en) | The method and device of waybill is provided | |
| US10032176B2 (en) | Real time statistics extraction from arbitrary advertising audiences | |
| CN111125574B (en) | Method and device for generating information | |
| US20120233173A1 (en) | Determining preferred categories based on user access attribute values | |
| US9052792B2 (en) | Inserting a search box into a mobile terminal dialog messaging protocol | |
| CN108510303A (en) | Advertisement placement method and device | |
| Wasserman et al. | Cross-evaluation of metrics to estimate the significance of creative works | |
| CN105956161A (en) | Information recommendation method and apparatus | |
| CN111506801A (en) | Sequencing method and device for sub-applications in application App | |
| CN108366096A (en) | A kind of information subscribing method, terminal and computer readable storage medium | |
| CN109388548A (en) | Method and apparatus for generating information | |
| CN108021673A (en) | A kind of user interest model generation method, position recommend method and computing device | |
| CN106062743A (en) | Systems and methods for keyword suggestion | |
| JP6976207B2 (en) | Information processing equipment, information processing methods, and programs | |
| WO2015042290A1 (en) | Identifying gaps in search results | |
| CN106709073A (en) | Browser notification pushing method and browser terminal | |
| CN106445971A (en) | Application recommendation method and system | |
| CN108416627A (en) | A kind of brand influence force monitoring method and system based on internet data | |
| CN106452809A (en) | Data processing method and device | |
| CN108694174B (en) | Content delivery data analysis method and device | |
| CN109190027A (en) | Multi-source recommended method, terminal, server, computer equipment, readable medium | |
| CN107798558A (en) | User behavior data analysis method and server |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220729 Address after: Room 801, 8th floor, No. 104, floors 1-19, building 2, yard 6, Jiuxianqiao Road, Chaoyang District, Beijing 100015 Patentee after: BEIJING QIHOO TECHNOLOGY Co.,Ltd. Address before: 100088 room 112, block D, 28 new street, new street, Xicheng District, Beijing (Desheng Park) Patentee before: BEIJING QIHOO TECHNOLOGY Co.,Ltd. Patentee before: Qizhi software (Beijing) Co.,Ltd. |