CN103544220B - Using recommendation method and apparatus - Google Patents
Using recommendation method and apparatus Download PDFInfo
- Publication number
- CN103544220B CN103544220B CN201310456253.5A CN201310456253A CN103544220B CN 103544220 B CN103544220 B CN 103544220B CN 201310456253 A CN201310456253 A CN 201310456253A CN 103544220 B CN103544220 B CN 103544220B
- Authority
- CN
- China
- Prior art keywords
- application
- app
- call relation
- compactness
- identities
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
- G06F16/24575—Query processing with adaptation to user needs using context
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of application and recommends method and apparatus, method to include：Obtain the application identities of the first application in terminal；According to the application identities of the first application, the application identities of multiple second applications with recommendation relation with the first application are searched in the local database；The application identities for meeting the second application for recommending requirement from extraction in the application identities of the multiple second applications are sent to first applies corresponding terminal, so that the second application that the terminal-pair is recommended is selected.The embodiment of the present invention is efficiently solved in prior art, and Mobile solution recommends method to recommend the technical problem of degraded performance.
Description
Technical field
The present invention relates to communication technical field, more particularly to a kind of application recommendation method and apparatus.
Background technology
With widely using for cell phone, Mobile solution is also occurred in that and is significantly increased.User is increasing
Find in Mobile solution that the process of an application for oneself needing also becomes difficult.It is that each Mobile solution is rationally accurately recommended
To user, Mobile solution field occurs in that many recommended technologies and method for application.
At present, the recommended technology in Mobile solution field is roughly divided into two classes：Traditional recommendation method is borrowed into application
On data set；The recommendation based on context is realized using the data that mobile device is collected.
Wherein, former approach is to have applied the classic algorithm transplanting of maturation to come in traditional recommendation field.For example,
Collaborative filtering based on user（User-based Collaborative Filtering, UCF）, collaborative filtering based on article
（Item-based Collaborative Filtering, ICF）And content-based recommendation（Content-based
Recommendation, CBR）.
Later approach is then using such as global positioning system（Global Positioning System, GPS）Sensor
To collect the information in the geographical position of user, improve the result of commending system according to these information.
But, all there are some defects in said method.Such as complete to apply the method recommended by transplanting classic algorithm, generally
The user that two possess similar sofware can be regarded as similar users, and then can will apply 1（App1）Recommend user 2
（User2）, will be using 2（App2）Recommend user 1（User1）, when App1 and App2 are map applications, then this recommendation
It is exactly the recommendation of the failure of repeatability.And be exactly geographical position with mobile device based on the essence of the recommendation of geographical location information
The collaborative filtering on basis is set to, its limitation is than larger.Also, when needing this as recommendation, premise must be in same place
There is the more than one mobile terminal for being provided with locating module, the suitability is poor.
The content of the invention
The present invention provides a kind of application and recommends method and apparatus, and to solve in prior art, Mobile solution recommends method
Recommend the technical problem of degraded performance.
On the one hand, the embodiment of the present invention provides one kind and applies recommendation method, including：
Obtain the application identities of the first application in terminal；
According to the application identities of the described first application, search in the local database and close with recommendation with the described first application
The application identities of multiple second applications of system；
The application identities that the second application for meeting recommendation requirement is extracted from the application identities of the plurality of second application are sent out
Described first is delivered to using the corresponding terminal, so that second application that the terminal-pair is recommended is selected.
On the other hand, the embodiment of the present invention provides one kind application recommendation apparatus, including：Acquisition module, searching modul and send out
Send module；
The acquisition module, for obtaining the application identities of the first application in terminal；
The searching modul, for according to described first application application identities, search in the local database with it is described
The application identities of multiple second applications of first application with recommendation relation；
The sending module, recommends require second for extracting to meet from the application identities of the plurality of second application
Using application identities send to described first using the corresponding terminal, for the terminal-pair recommend described second should
With being selected.
Method and apparatus is recommended in the application that the present invention is provided, and is getting in terminal after the application identities of the first application, root
According to the application identities of first application, searching in the local database should with first application with recommendation relation multiple second
Application identities；The application mark for meeting the second application for recommending to require is extracted from the application identities of the plurality of second application
Know to send to first and apply corresponding terminal, so that above-mentioned second application that the terminal-pair is recommended is selected, improve application
The performance of recommendation.
Description of the drawings
The flow chart of application recommendation method one embodiment that Fig. 1 is provided for the present invention；
The flow chart of application recommendation method another embodiment that Fig. 2 is provided for the present invention；
The structural representation of application recommendation apparatus one embodiment that Fig. 3 is provided for the present invention；
The structural representation of application recommendation apparatus another embodiment that Fig. 4 is provided for the present invention.
Specific embodiment
The flow chart of application recommendation method one embodiment that Fig. 1 is provided for the present invention.As shown in figure 1, following steps
Executive agent can be the network equipment in network, server, or the module on being integrated in the network equipment or server, core
Piece etc..As shown in figure 1, this is specifically included using recommendation method：
S101, obtains the application identities of the first application in terminal；
Wherein, the terminal that this programme is related to can be the various electronic device terminals such as mobile phone, notebook computer, Ipad.Should
The first application in terminal can be specifically that the various networks such as QQ chat applications, Renren Network client software, microblogging, wechat should
With.And the first application identities applied unique mark then for sign first application, can be specifically answering for first application
With title etc..When user downloads in application market and installs these in the terminal and apply, or user using these using soft
During part, can be known which application be currently installed in terminal by technologies such as network monitors, and obtain these and be answered
Application identities.
S102, according to the application identities of the first application, is searched with the first application in the local database with recommendation relation
It is multiple second application application identities；
After the application identities of the in terminal first application are got, according to the application identities, look in the local database
Look for the application identities of several the second applications with recommendation relation with first application.Wherein, the local data base is deposited in advance
Contain carries out the application identities of some second applications using recommendation for the first application.
S103, the application identities that the second application for meeting recommendation requirement is extracted from the application identities of the multiple second applications send out
First is delivered to using corresponding terminal, so that the second application that the terminal-pair is recommended is selected；
After the application identities of the in terminal first application are got, answering for the first application can be directed in the local database
In application identities with several the second applications of mark correspondence storage, extract and meet the second application for recommending to require.The recommendation
Requirement can be configured by terminal by user, or directly be arranged by network management personnel, can be specifically to recommend to answer
Number, size of application file etc..Extract after meeting the second application for recommending to require, can specifically by second application
The information such as application identities, such as Apply Names, the download link address of application is sent to first to be applied in corresponding terminal, for end
The second application to recommending is held to be selected and installed.
The application recommendation method that the present invention is provided, after the application identities for getting in terminal the first application, according to this
The application identities of one application, in the local database lookup and first application have answering for multiple second applications of recommendation relation
With mark；The application identities that the second application for meeting recommendation requirement is extracted from the application identities of the plurality of second application send
Corresponding terminal is applied to first, so that above-mentioned second application that the terminal-pair is recommended is selected, is improve using recommendation
Performance.
The flow chart of application recommendation method another embodiment that Fig. 2 is provided for the present invention, is embodiment as shown in Figure 1
A kind of specific implementation of method, the method are specifically included：
S201, obtains the application identities of the first application in terminal；The concrete implementation procedure of the step can be found in the phase of step 101
Answer content.
S202, in calculating the first application and the multiple second applications, each second applies the similarity on theme, and/or,
Compactness on call relation；
Wherein, the similarity on theme, for characterize first application and second application between application type, field,
Similarity present on the themes such as function.Such as microblogging, wechat, QQ chat applications etc. belong to the application of instant messaging class, because
This their similarity on theme is higher；And between wechat and tiger map software, each belonging to functional area difference compared with
Greatly, thus their similarities on theme are relatively low.
And the compactness on call relation, then apply and the second application between with the presence or absence of more tight for characterizing first
Call relation.For example, there is certain call relation with the second application in the first application, and this call relation should relative to other
With even closer with the call relation of the first application, more it is difficult to by other application in other words replace which to apply with first
Position on this call relation.So, it is believed that second application is compared compared with other application, which is called with the first application
Relation is more tight, there is a kind of irreplaceability.Now, the first application is had with which and this more closely calls pass by we
Irreplaceability between second application of system is defined as the first application and applies tight on corresponding call relation with second
Degree.For example, this more close irreplaceability is there is between shopping website and payment software, is being adjusted between both applications
With higher compared with its compactness between other application and shopping website in relation.
By the above-mentioned similarity on theme, and the judging basis of the compactness on call relation, calculating first should
With with the multiple second applications in each second apply the similarity on theme, and/or, the compactness on call relation.This
Embodiment gives a kind of calculating two similarities applied on theme, and one kind of the compactness on call relation is concrete
Calculation is as follows.
1. in calculating the first application and the multiple second applications, each second applies the similarity on theme, including：
11, obtain the description information comprising the above-mentioned first application with the multiple second multiple applications applied；
Wherein, the description information of application can be understood as the function with regard to a certain application, art, operating feature etc.
The information of comprehensive description is carried out to the application.Most intuitively then can using the page-describing information of a certain application in application market as
The foregoing description information of the application.Calculated most by Topic Similarity in above-mentioned first application and the second application as the present embodiment
Two object applications for having recommendation relation are determined eventually.
12, the description information of multiple applications is divided into by several notional words by participle instrument；
By existing participle instrument in prior art, will get many with what second applied comprising the first application
The description information of individual application carries out notional word division, can obtain the corresponding notional word set of description information applied with regard to each.
13, for each notional word K that each includes using the description information of j in multiple applicationsi, according to single text word
Frequently-inverse document frequency（term frequency–inverse document frequency）TF-IDF algorithms calculate notional word Ki
Weight w in the description information using ji,j；
Wherein, with regard to TF-IDF algorithms, the principle of the algorithm is retouched to the related of calculating process in can be found in prior art
State, therefore not to repeat here.
In this programme, basis firstCalculate notional word KiDan Wen in the description information of the application j
This word frequency TFi,j；Wherein, the fi,jFor notional word KiThe number of times occurred in the description information using j, the Σzfz,jIt is using j's
All notional word K in description informationzThe total degree sum of appearance.
According toCalculate notional word KiInverse document frequency IDF in the description information of above-mentioned multiple applicationsi；Its
In, N is the numerical value of the fixed number, niFor notional word K is included in the description information of N number of applicationiApplication number.
According to wi,j=TFi,j×IDFiCalculate notional word KiWeight w in the description information using ji,j；Wherein, TFi,jFor reality
Word KiAbove-mentioned single text word frequency in the description information of the application j, IDFiFor the notional word KiIn retouching for above-mentioned multiple applications
State the inverse document frequency IDF in informationi。
14, according toBuild the characteristic vector of the description information using j
With using each notional word K in jiWeight w in the description information using ji,jAs characteristic item, build using j's
The characteristic vector of description information
15, according toDetermine first application with it is multiple
In second application, each second applies the similarity Similarity (App on theme1,App2)；
For two similarities applied on theme, folder cosine of an angle of this programme from the characteristic vector of two applications
Value is being judged.When the folder cosine of an angle of the characteristic vector of two applications is equal to 1, the angle of corresponding angle is 0, is shown
Two identical using theme；When the folder cosine of an angle of the characteristic vector of two applications is less, the angle of corresponding angle is got over
Greatly, show that the theme of two applications is more uncorrelated.
2. in calculating the first application and the multiple second applications, each second applies the compactness on call relation, including：
21, obtain and call configuration file with the multiple second multiple applications applied comprising the first application；
By the binary file after the code source file for extracting the installation kit of each application, that is, program code compiling,
Then the configuration file that decompiling can be obtained by the code source file and correlation of this application program is carried out to which.
In the platform of such as android system, there is provided the mechanism of a kind of entitled " Intent ".The Intent is responsible for correspondence
It is related to data, additional data with the action of middle once-through operation, action to be described, android system is then according to this Intent's
Description, is responsible for finding corresponding component or application, and Intent is passed to it, to complete calling between component or application.
Intent not only can complete calling between different components in an application program apk, it is also possible to complete the tune using between
With.Mechanism for resolving in this programme using Intent searches registered all Intent in AndroidManifest.xml
Filter files and its defined in Intent, and the institute between the characterized application of Intent that these are defined and correlation
What the message file for having call relation was respectively applied in being defined as this step calls configuration file.If the Intent of two applications is obtained
The matching arrived in relation, we then think the annexation that Intent is there is between the two applications, i.e., in the present embodiment
Call relation between two applications being related to.
22, for each calls configuration file using j in multiple applications, call configuration text in the plurality of application
Extract and the application using j with call relation and corresponding call relation type in part；
For in above-mentioned multiple applications each call configuration file using j, match somebody with somebody in all calling of the plurality of application
Put in file, extract the application that there is call relation using j with this, and specific call relation type.Judge that two are answered
Whether there is between call relation, configuration file above-mentioned can be called using corresponding by each（Intent）, by content
Match to judge.The call relation in the platform of android system, between common application is listed in table 1.
Call relation between the application of table 1
23, for each and the call relation type having using j, calculate the call relation type and answer the plurality of
With the number of times occurred between interior each application；
The call relation between each application from table 1 would know that the call relation between each two application all differs
Sample, and the degree of contact between corresponding two applications for embodying of every kind of call relation between each application is also different
's.In order to more intuitively show the contact between above-mentioned application, need formally to assess the contact between them
It is strong and weak.
First, for it is above-mentioned it is multiple application in each apply j, count this using j with it is the plurality of application in other
The species of the call relation type having using between, and it is more at this to calculate the call relation type with regard to every kind of call relation type
It is individual using the total degree that occurs between interior each application.
24, according toCalculate first to apply
With one second compactness weight Weight (e (t, the App applied on call relation of the type for t1,App2)), wherein, institute
State E (e (t, App1,App2)) it is the call relation of t types existed between the first application and one second application the plurality of
Using the number of times occurred between interior each application；
The call relation of a certain type for existing between two applications, calculates the call relation and applies it at the two
Between call relation on compactness weight, wherein, E (e (t, App1,App2)) for, in two applications, the first application should with second
The total degree that the call relation of the t types existed between occurs between each application in above-mentioned fixed number application.
25, according to Dependency (App1,App2)=ΣTWeight(e(t,App1,App2))（4）, determine the first application
Compactness Dependency (the App on call relation is applied with one second1,App2)；
Compactness weight on the call relation for type used between above-mentioned first application and the second application is entered
Row summation, that is, obtain the compactness on call relation between two applications.Wherein, Weight (e (t, the App1,App2))
For the first application and the one second compactness weight applied on call relation of the type for t, the T is the first application and
The set of all call relation types existed between individual second application.
S203, applies the similarity on theme according to first, and/or, the size of the compactness on call relation
To multiple second application be ranked up, and according to ranking results by second application application identities with first application application identities
Correspondence storage is in the local database；
Similarity size on theme is applied with first by calculated multiple second applications, or calling pass
The compactness size fastened, or similarity and compactness consider, and above-mentioned multiple second applications are ranked up, and according to
It is stored in the application identities of the second application corresponding with the application identities of the first application in the local data base according to ranking results.
This gives considering based on similarity and compactness, realizes the concrete reality being ranked up to the multiple second applications
Apply mode as follows：
Extract each second application the similarity on theme, and the compactness on call relation are applied with first,
And pass through
Calculate the recommendation CombineValue (App of each second application and the first application1,App2), and will be the plurality of
Order of second application according to recommendation from big to small is ranked up；Wherein, the Similarity (App1,App2) for first
The similarity on theme, the Dependency (App are applied using with second1,App2) apply with second for the first application
Compactness on call relation.
S204, according to the application identities of the first application, is searched with the first application in the local database with recommendation relation
It is multiple second application application identities；The concrete implementation procedure of the step can be found in the corresponding contents of step 102.
S205, the application identities that the second application for meeting recommendation requirement is extracted from the application identities of the multiple second applications send out
First is delivered to using corresponding terminal, so that the second application that the terminal-pair is recommended is selected；The concrete implementation procedure of the step
Can be found in the corresponding contents of step 103.
Specifically, based on above by calculated each second application and the first recommendation applied
CombineValue(App1,App2), the method are ranked up by several second applications.The embodiment of the present invention gives step
A kind of rapid 103 implementation：The application of top n second in multiple second applications after will sorting is sent to the described first application
Corresponding terminal, so that the application that terminal-pair is recommended is selected.Wherein, N is positive integer.
The application recommendation method that the present invention is provided, after the application identities for getting in terminal the first application, according to this
The application identities of one application, in the local database lookup and first application have answering for multiple second applications of recommendation relation
With mark；The application identities that the second application for meeting recommendation requirement is extracted from the application identities of the plurality of second application send
Corresponding terminal is applied to first, so that above-mentioned second application that the terminal-pair is recommended is selected, is improve using recommendation
Performance.
One of ordinary skill in the art will appreciate that：Realize that all or part of step of above-mentioned each method embodiment can be led to
Cross the related hardware of programmed instruction to complete.Aforesaid program can be stored in a computer read/write memory medium.The journey
Sequence upon execution, performs the step of including above-mentioned each method embodiment；And aforesaid storage medium includes：ROM, RAM, magnetic disc or
Person's CD etc. is various can be with the medium of store program codes.
Fig. 3 provides the structural representation using recommendation apparatus one embodiment for the present invention.The structural representation can be held
Row is such as the step of application recommendation method in Fig. 1 embodiments.As shown in figure 4, this includes using recommendation apparatus：Acquisition module 31,
Searching modul 32, sending module 33；Wherein：
Acquisition module 31, for obtaining the application identities of the first application in terminal；
Searching modul 32, for the application identities according to the first application, is searched and the first application tool in the local database
There are the application identities of multiple second applications of recommendation relation；
Sending module 33, recommends the require second application for extracting to meet from the application identities of the multiple second applications
Application identities send to first and apply corresponding terminal, so that the second application that terminal-pair is recommended is selected.
Specifically, performing corresponding application using recommendation apparatus shown in the present embodiment recommends the detailed process of method as follows：
Acquisition module 31 obtains the application identities of the first application in terminal；Application mark of the searching modul 32 according to the first application
Know, search the application identities of multiple second applications with recommendation relation with the first application in the local database；Sending module
33 application identities that the second application for meeting recommendation requirement is extracted from the application identities of the multiple second applications send should to first
Corresponding terminal is used, so that the second application that terminal-pair is recommended is selected.
The application recommendation apparatus that the present invention is provided, after the application identities for getting in terminal the first application, according to this
The application identities of one application, in the local database lookup and first application have answering for multiple second applications of recommendation relation
With mark；The application identities that the second application for meeting recommendation requirement is extracted from the application identities of the plurality of second application send
Corresponding terminal is applied to first, so that above-mentioned second application that the terminal-pair is recommended is selected, is improve using recommendation
Performance.
The structural representation of application recommendation apparatus another embodiment that Fig. 4 is provided for the present invention, is enforcement as shown in Figure 3
A kind of specific implementation of the structural representation of example, on the basis of the 26S Proteasome Structure and Function of embodiment as indicated at 3, also includes：
Processing module 34；
The processing module 34, second applies the phase on theme for calculating in the first application and the multiple second applications each
Like degree, and/or, the compactness on call relation；
The similarity on theme is applied according to first, and/or, the size of the compactness on call relation is to multiple
Second application is ranked up, and deposits the application identities of the second application corresponding with the application identities of the first application according to ranking results
Storage is in the local data base.
Specifically, the application recommendation apparatus that the present embodiment is provided can be used to perform shown in above-mentioned all embodiments of the method
Using the flow process for recommending method, its concrete operating principle is repeated no more, and refers to the description of embodiment of the method.
The application recommendation apparatus that the present invention is provided, after the application identities for getting in terminal the first application, according to this
The application identities of one application, in the local database lookup and first application have answering for multiple second applications of recommendation relation
With mark；The application identities that the second application for meeting recommendation requirement is extracted from the application identities of the plurality of second application send
Corresponding terminal is applied to first, so that above-mentioned second application that the terminal-pair is recommended is selected, is improve using recommendation
Performance.
Finally it should be noted that：Various embodiments above only to illustrate technical scheme, rather than a limitation；To the greatest extent
Pipe has been described in detail to the present invention with reference to foregoing embodiments, it will be understood by those within the art that：Its according to
So the technical scheme described in foregoing embodiments can be modified, or which part or all technical characteristic are entered
Row equivalent；And these modifications or replacement, do not make the essence of appropriate technical solution depart from various embodiments of the present invention technology
The scope of scheme.
Claims (4)
1. it is a kind of to apply recommendation method, it is characterised in that to include：
Obtain the application identities of the first application in terminal；
Calculate in first application and the multiple second applications each and second apply similarity on theme and in call relation
On compactness, according to the size of the similarity on theme and the compactness on call relation is applied with described first to institute
State multiple second applications to be ranked up, or, calculate in first application and the multiple second applications each and second applying and calling
Compactness in relation, applies to the plurality of second according to the size of the compactness on call relation is applied with described first
It is ranked up, and is stored in the application identities of the described second application corresponding with the application identities of the first application according to ranking results
In local data base, wherein, the similarity is calculated according to the corresponding characteristic vector of description information of application, described tight
Spend and calculated according to the information that configuration file is included of calling of application；
According to the application identities of the described first application, search in the local database and there is recommendation relation with the described first application
The application identities of multiple second applications；
From the plurality of second application application identities in extract meet recommend require second application application identities send to
Described first applies the corresponding terminal, so that second application that the terminal-pair is recommended is selected；
In calculating first application and the plurality of second application, each second applies the similarity on theme, including：
Obtain the description information comprising the described first application with the plurality of second multiple applications applied；
The description information of the plurality of application is divided into by several notional words by participle instrument；
For each notional word K that each includes using the description information of j in the plurality of applicationi, according to single text word frequency-
Inverse document frequency TF-IDF algorithms calculate the notional word KiWeight w in the description information of the application ji,j；
According toBuild the characteristic vector of the description information of the application j
According toDetermine that first application should with the plurality of second
With in, each second applies the similarity Similarity (App on theme1,App2)；
In calculating first application and the plurality of second application, each second applies the compactness on call relation, wraps
Include：
Obtain and call configuration file with the plurality of second multiple application applied comprising the described first application；
For each calls configuration file using j in the plurality of application, configuration file is called in the plurality of application
Middle extraction has the application of call relation and corresponding call relation type with the application j；
For the call relation type that each is had with the application j, the call relation type is calculated in the plurality of application
The number of times occurred between interior each application；
According toFirst application is calculated with one second
Apply compactness weight Weight (e (t, the App on call relation of the type for t1,App2)), wherein, the E (e (t,
App1,App2)) it is the call relation of the t types existed between first application and one second application described
It is multiple using the number of times that occurs between interior each application；
According to Dependency (App1,App2)=∑TWeight(e(t,App1,App2)), determine first application with one
Second applies the compactness Dependency (App on call relation1,App2)；Wherein, Weight (e (t, the App1,
App2)) it is first application and the one second compactness weight applied on call relation of the type for t, the T is institute
State the set of all call relation types existed between the first application and one second application.
2. method according to claim 1, it is characterised in that described to apply similar on theme according to described first
Degree, and, the size of the compactness on call relation is ranked up to the plurality of second application, including：
Extract each described second application the similarity on theme is applied with described first, and it is tight on call relation
Degree, and pass through
Calculate the recommendation CombineValue (App of each described second application and the described first application1,App2), and will be described
Order of multiple second applications according to recommendation from big to small is ranked up；Wherein, the Similarity (App1,App2) be
First application applies the similarity on theme, the Dependency (App with described second1,App2) for described first
The compactness on call relation is applied using with described second.
3. method according to claim 2, it is characterised in that the recommendation requires to include recommendation number N, and N is positive integer；
The extraction from the plurality of second application meets the second application transmission for recommending to require to the described first application correspondence
The terminal, for the terminal-pair recommend it is described second application selected, including：
The application of top n second in multiple second applications after by sequence sends to described first and applies corresponding terminal, for end
Hold and the application recommended is selected.
4. one kind application recommendation apparatus, it is characterised in that include：Acquisition module, searching modul and sending module；
The acquisition module, for obtaining the application identities of the first application in terminal；
Processing module, second applies the similarity on theme for calculating in first application and the multiple second applications each
With the compactness on call relation, according to the similarity applied with described first on theme and tight on call relation
The size of degree is ranked up to the plurality of second application, or, every with the multiple second applications for calculating first application
Individual second applies the compactness on call relation, according to the size applied with described first in the compactness on call relation
The plurality of second application is ranked up, and should with described first by the application identities of the described second application according to ranking results
In the local database, the similarity is according to the corresponding characteristic vector of description information applied for the storage of application identities correspondence
Calculated, the compactness is calculated according to the information that configuration file is included of calling of application；The searching modul, is used for
According to the application identities of the described first application, search in the local database multiple with recommendation relation with the described first application
The application identities of the second application；
The sending module, recommends the require second application for extracting to meet from the application identities of the plurality of second application
Application identities send to described first using the corresponding terminal, so that second application that the terminal-pair is recommended is entered
Row is selected；
The processing module calculates in first application and the plurality of second application each and second applies the phase on theme
Seemingly spend, including：
Obtain the description information comprising the described first application with the plurality of second multiple applications applied；
The description information of the plurality of application is divided into by several notional words by participle instrument；
For each notional word K that each includes using the description information of j in the plurality of applicationi, according to single text word frequency-
Inverse document frequency TF-IDF algorithms calculate the notional word KiWeight w in the description information of the application jI, j；
According toBuild the characteristic vector of the description information of the application j
According toDetermine that first application should with the plurality of second
With in, each second applies the similarity Similarity (App on theme1,App2)；
The processing module calculates in first application and the plurality of second application each and second applies on call relation
Compactness, including：
Obtain and call configuration file with the plurality of second multiple application applied comprising the described first application；
For each calls configuration file using j in the plurality of application, configuration file is called in the plurality of application
Middle extraction has the application of call relation and corresponding call relation type with the application j；
For the call relation type that each is had with the application j, the call relation type is calculated in the plurality of application
The number of times occurred between interior each application；
According toFirst application is calculated with one second
Apply compactness weight Weight (e (t, the App on call relation of the type for t1,App2)), wherein, the E (e (t,
App1,App2)) it is the call relation of the t types existed between first application and one second application described
It is multiple using the number of times that occurs between interior each application；
According to Dependency (App1,App2)=∑TWeight(e(t,App1,App2)), determine first application with one
Second applies the compactness Dependency (App on call relation1,App2)；Wherein, Weight (e (t, the App1,
App2)) it is first application and the one second compactness weight applied on call relation of the type for t, the T is institute
State the set of all call relation types existed between the first application and one second application.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310456253.5A CN103544220B (en) | 2013-09-29 | 2013-09-29 | Using recommendation method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310456253.5A CN103544220B (en) | 2013-09-29 | 2013-09-29 | Using recommendation method and apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103544220A CN103544220A (en) | 2014-01-29 |
| CN103544220B true CN103544220B (en) | 2017-04-05 |
Family
ID=49967672
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310456253.5A Active CN103544220B (en) | 2013-09-29 | 2013-09-29 | Using recommendation method and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103544220B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104866505B (en) * | 2014-02-25 | 2021-04-06 | 腾讯科技（深圳）有限公司 | Application recommendation method and device |
| CN104951324A (en) * | 2014-03-28 | 2015-09-30 | 腾讯科技（深圳）有限公司 | Method and device for acquiring application calling relation table |
| CN105446994A (en) * | 2014-07-11 | 2016-03-30 | 华为技术有限公司 | Service recommendation method and device with intelligent assistant |
| CN105893390B (en) * | 2015-01-26 | 2021-06-22 | 北京搜狗科技发展有限公司 | Application processing method and electronic equipment |
| CN104935651A (en) * | 2015-06-03 | 2015-09-23 | 魅族科技（中国）有限公司 | Content recommending method and terminal |
| CN106302606B (en) * | 2015-06-08 | 2019-11-29 | ***通信集团湖南有限公司 | Across the application access method and device of one kind |
| CN105630977B (en) * | 2015-12-25 | 2019-03-26 | 腾讯科技（深圳）有限公司 | Application program recommended method, apparatus and system |
| CN107402879B (en) * | 2016-05-20 | 2020-11-06 | 阿里巴巴集团控股有限公司 | Method and device for determining call relation between applications |
| CN107133296B (en) * | 2017-04-26 | 2020-08-21 | 南京心视窗信息科技有限公司 | Application program recommendation method and device and computer readable storage medium |
| CN107317930B (en) * | 2017-06-26 | 2020-12-15 | 南京沄海区块链科技有限公司 | Desktop icon layout method and device and computer readable storage medium |
| CN108182201B (en) * | 2017-11-29 | 2020-06-30 | 有米科技股份有限公司 | Application expansion method and device based on key keywords |
| CN108769125B (en) * | 2018-04-28 | 2021-08-17 | 阿里巴巴（中国）有限公司 | Application recommendation method and device, storage medium and computer equipment |
| CN112698932A (en) * | 2021-03-23 | 2021-04-23 | 中国信息通信研究院 | Industrial application program calling method and device, computer equipment and storage medium |
| CN114443967B (en) * | 2022-04-08 | 2022-07-08 | 北京并行科技股份有限公司 | Similar application recommendation method, computing device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101441662A (en) * | 2008-11-28 | 2009-05-27 | 北京交通大学 | Topic information acquisition method based on network topology |
| CN102360364A (en) * | 2011-09-30 | 2012-02-22 | 奇智软件（北京）有限公司 | Automatic application recommendation method and device |
| CN102880501A (en) * | 2012-07-24 | 2013-01-16 | 北京奇虎科技有限公司 | Realizing method, device and system for recommending applications |
-
2013
- 2013-09-29 CN CN201310456253.5A patent/CN103544220B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101441662A (en) * | 2008-11-28 | 2009-05-27 | 北京交通大学 | Topic information acquisition method based on network topology |
| CN102360364A (en) * | 2011-09-30 | 2012-02-22 | 奇智软件（北京）有限公司 | Automatic application recommendation method and device |
| CN102880501A (en) * | 2012-07-24 | 2013-01-16 | 北京奇虎科技有限公司 | Realizing method, device and system for recommending applications |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103544220A (en) | 2014-01-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103544220B (en) | Using recommendation method and apparatus | |
| US11218505B2 (en) | Facilitating cross-platform content access | |
| CN107341187B (en) | Search processing method, device, equipment and computer storage medium | |
| CN103530751B (en) | The method and device of waybill is provided | |
| CN111046237B (en) | User behavior data processing method and device, electronic equipment and readable medium | |
| CN110262851B (en) | Interface configuration method and device | |
| Costa et al. | Emotion-based recommender system for overcoming the problem of information overload | |
| CN104750760A (en) | Application software recommending method and device | |
| CN113435862B (en) | Bill processing method and device based on mailbox | |
| JP2022096632A (en) | Computer-implemented method, computer system, and computer program (ranking datasets based on data attributes) | |
| CN111160029B (en) | Information processing method and device, electronic equipment and computer readable storage medium | |
| CN113221036B (en) | Method and device for processing electronic bill mail | |
| US20220179916A1 (en) | Computerized grouping of news articles by activity and associated phase of focus | |
| US20080162503A1 (en) | Application of community-defined descriptors to mobile content | |
| US9251273B2 (en) | Delivering a filtered search result | |
| CN106294417A (en) | A kind of data reordering method, device and electronic equipment | |
| CN104778233A (en) | Searching method and device based on click rate | |
| Kuck et al. | Context-sensitive service discovery meets information retrieval | |
| CN104881308B (en) | A kind of processing method and processing device of application program | |
| US9843635B2 (en) | Data replication among portable electronic devices | |
| CN105378710B (en) | Identification search matching and the search result through changing | |
| Guo et al. | A graph-based push service platform | |
| CN104572945A (en) | File search method and device based on cloud storage space | |
| CN107544685A (en) | Information-pushing method and device | |
| US9727652B2 (en) | Utilizing dependency among internet search results |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |