CN106682056B - The determination method, apparatus and system of correlation between different application software - Google Patents
The determination method, apparatus and system of correlation between different application software Download PDFInfo
- Publication number
- CN106682056B CN106682056B CN201610559788.9A CN201610559788A CN106682056B CN 106682056 B CN106682056 B CN 106682056B CN 201610559788 A CN201610559788 A CN 201610559788A CN 106682056 B CN106682056 B CN 106682056B
- Authority
- CN
- China
- Prior art keywords
- application software
- pair
- centering
- users
- correlation
- 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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Stored Programmes (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The determination method, apparatus and system of correlation between a kind of different application software are disclosed in the embodiment of the present invention.Wherein, method includes:The application software for obtaining multiple users uses list;Determine multiple application software pair;For each application software pair, list is used according to acquired application software, the concurrent users number of the application software pair is determined from the multiple user, and two respective numbers of users of application software of the application software centering, and the correlation characterization index being calculated between two application software of the application software centering in turn.Technical solution in the embodiment of the present invention can objectively and accurately find out the correlation between application software, and can quantify to this relationship.
Description
Technical field
This application involves Internet technical field, the determination side of correlation between especially a kind of different application software
Method, apparatus and system.
Background technique
With the popularization of intelligent terminals, for being mounted on the application software used on intelligent terminal, such as it is mounted on intelligent hand
Application software (APP, Application Program) on machine, type it is also more and more abundant.It can between different application software
With certain correlation, such as promote relationship and competitive relation.By taking APP as an example, if APP1 is promotion relationship to APP2,
Then mean to have used the user of APP1 can also be inclined to use APP2;If APP1 is competitive relation to APP2, mean to make
With the user of APP1, it is intended to not go using APP2.
There is the correlation between different application software, it is soft can be used to recall relevant application in recommender system
Part, to improve the recommendation efficiency of recommender system.For example, by taking mobile phone house keeper as an example, in the APP software electrical market of mobile phone house keeper
In, the favorite top n APP of the user can be recommended to show position to user, but the use can not usually be known for the APP of magnanimity
Family can only know that the user to the fancy grade of certain APP, at this moment can be liked using user to the fancy grade of all APP
The correlation of good APP and other APP finally obtain to judge the user to the fancy grade of other APP to each APP
Hobby value, just can determine that top n APP by being ranked up, i.e., recall it from several APP of user preferences and like multiple
APP.For example, if knowing, a user likes that value is very high to ' Baidu's take-away ', and ' Baidu's take-away ' and ' being hungry ' it is mutual
Promotion relationship is obvious, it can be considered that the user also can be relatively high to the hobby value being hungry.
In addition, there is the correlation between different application software, APP can also be associated in advertisement delivery system
It launches, to improve the dispensing efficiency of advertisement delivery system.For example, for above example, if to be launched to ' Baidu's take-away '
Advertisement then can choose installation or the user of ' being hungry ' used to launch.
It in addition, there is the correlation between different application software, can also assist carrying out competing product discovery, be used for company
Reference when layer decision-making level and other companies cooperate uses, and improves the efficiency of decision-making.
It is at present usually the registration simply from two APP functionally to determine the correlation between two APP
Or classification situation is judged.For example, an example after being judged according to this judgment method is:Alipay and Taobao
To promote relationship, Taobao and Jingdone district are competitive relation, and mobile phone house keeper and 360 security guards are competitive relation etc..
But in fact, it is above-mentioned competitive relation that the Taobao and Jingdone district in above-mentioned example, which are not, usually using the use of Taobao
Family also will use Jingdone district.Therefore, the APP that single function or classifieds website (such as pea pods or reference treasured) from APP provides divides
If category information is judged, accuracy not can guarantee;In addition, the judging result can not also quantify, that is, promote or compete to arrive
There are much degree at bottom, unknown.
Summary of the invention
In view of this, on the one hand providing a kind of determination side of correlation between different application software in the embodiment of the present invention
On the other hand method provides a kind of determination device and system of correlation between different application software, objectively and accurately to look for
Correlation between application software out, and this relationship can be quantified.
The determination method of correlation between a kind of different application software provided in the embodiment of the present invention, including:
The application software for obtaining M user uses list;M is the positive integer more than or equal to 2;Determine N number of application software
It is right;N is the positive integer more than or equal to 1;For each application software pair, list is used according to acquired application software, from
Two applications of concurrent users number and the application software centering that the application software pair is determined in the M user are soft
The respective number of users of part;For each application software pair, according to the concurrent users number of the application software pair and the application
Two respective numbers of users of application software of software centering, are calculated between two application software of the application software centering
Correlation characterize index.
The determining device of correlation between a kind of different application software provided in the embodiment of the present invention, including:List
Module is obtained, the application software for obtaining M user uses list;M is the positive integer more than or equal to 2;Software is to determination
Module, for determining N number of application software pair;N is the positive integer more than or equal to 1;Number of users determining module, for for each
Application software pair uses list according to acquired application software, and the determination application software is to being total to from the M user
Deposit two respective numbers of users of application software of number of users and the application software centering;Index computing module, for being directed to
Each application software pair, according to the concurrent users number of the application software pair and two applications of the application software centering
The respective number of users of software, the correlation characterization being calculated between two application software of the application software centering refer to
Mark.
The determination system of correlation between a kind of different application software provided in the embodiment of the present invention, including:
The determining device of correlation between M intelligent terminal and a kind of above-mentioned different application software;Wherein, the M
Each intelligent terminal in a intelligent terminal is used to the application software of own user being supplied to described device using list;M is
Positive integer more than or equal to 2.
As it can be seen that the application software by obtaining multiple users uses list, and according to acquired in the embodiment of the present invention
Application software list determines multiple application software pair;For each application software pair, column are used according to acquired application software
Table determines that the concurrent users number of the application software pair and two of the application software centering are answered from the M user
With the respective number of users of software, and then the correlation table being calculated between two application software of the application software centering
Levy index, it is seen then that the characteristic index accurately has found the correlation between application software based on objective data, and
And it is presented in a manner of numerical value, the quantization to correlation between two application software is realized, so that not only may determine that
Correlation between two application software is promotion relationship or competitive relation, and can quantify its journey for promoting or competing
Degree.
Detailed description of the invention
For the clearer technical solution illustrated in the embodiment of the present invention, will make below to required in embodiment description
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing.Wherein,
The exemplary flow of Fig. 1 determination method of correlation between different application software a kind of in the embodiment of the present invention
Figure;
Fig. 2 is the concurrent users number of application software pair in the embodiment of the present invention, the first application software of application software centering
Correlation schematic diagram between the number of users three of the second application software of number of users and application software centering;
The calculating knot of Android editions synchronous correlations between assistant and some softwares in Fig. 3 one example of the present invention
Fruit schematic diagram;
The schematic diagram of calculation result of the correlation between money and valuables and some softwares is dug in Fig. 4 one example of the present invention;
The exemplary structure of Fig. 5 determining device of correlation between different application software a kind of in the embodiment of the present invention
Figure;
Fig. 6 is the structural schematic diagram of comprehensive score determining module shown in Fig. 5 in an example of the invention;
Fig. 7 is the structural schematic diagram of comprehensive score determining module shown in Fig. 5 in another example of the invention;
Fig. 8 is a kind of structural schematic diagram of server in the embodiment of the present invention;
The exemplary structure of Fig. 9 determination system of correlation between different application software a kind of in the embodiment of the present invention
Figure.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, by the following examples to of the invention further detailed
It describes in detail bright.
It is succinct and intuitive in order to what is described, hereafter by describing several representative embodiments come to side of the invention
Case is illustrated.A large amount of details is only used for helping to understand the solution of the present invention in embodiment.However, it will be apparent that of the invention
Technical solution can be not limited to these details when realizing.In order to avoid unnecessarily having obscured the solution of the present invention, Yi Xieshi
It applies mode not described meticulously, but only gives frame.Hereinafter, " comprising " refers to " including but not limited to ", " root
According to ... " refer to " according at least to ..., but be not limited to according only to ... ".Due to the speech habits of Chinese, hereinafter without spy
When not pointing out the quantity of an ingredient, it is meant that the ingredient is either one or more, or can be regarded as at least one.
The exemplary flow of Fig. 1 determination method of correlation between different application software a kind of in the embodiment of the present invention
Figure.Method in the present embodiment can be applied in server, server cluster or cloud computing service center.As shown in Figure 1, should
Method may include following steps:
Step 101, the application software for obtaining M user uses list.Wherein, M is the positive integer more than or equal to 2.It is real
In the application of border, M can be the quantity of hundred, thousand, ten thousand ranks.
In this step, obtain M user application software use list mode can there are many.For example, can be by each
The intelligent terminal of user uploads the application software of the user using list, then in this step, can directly receive each intelligence
The application software for the user that energy terminal uploads uses list.Alternatively, can also be obtained from each application software server in this step
The application software of each user is taken to use list.Or can also be used in this step other modes obtain each user using soft
Part uses list.
Wherein, application software can be the application software list of user's actual use using list, be also possible to user's peace
Application software list on intelligent terminal, is not limited thereof herein.
Below in case where intelligent terminal is mobile phone terminal, it is assumed that the application software (APP) of reporting of user uses column
Table is that user has used those softwares daily, and data are as follows:
Format:User 1 (IMEI1), APP1, APP2, APP3...
User 2 (IMEI1), APP1, APP4...
User 3 (IMEI1), APP2, APP3, APP4, APP5...
Wherein, according to the explanation in wikipedia, IMEI (International Mobile Equipment
It Identity) is international mobile equipment identification number, i.e., usually said handset serial, mobile phone " string number ", in mobile electricity
The mobile communication devices such as each independent mobile phone are identified in telephone network.
Step 102, N number of application software pair is determined.Wherein, N is the positive integer more than or equal to 1.
In this step, determine N number of application software pair mode can there are many.For example, can according to user demand or its
He is tactful, predefines application software pair to be measured.Alternatively, all application software acquired in step 101 can also be made
It is matched two-by-two with the application software in list, obtains each application software pair.Alternatively, can also be according in step 101
Acquired application software uses list, determines the N number of application software pair for meeting setting demand.In this step, if predefining
If application software pair to be measured, then there is no absolute precedence relationship between step 101 and step 102, step can be first carried out
102 execute step 101 or two steps are performed simultaneously again.If determining application software pair according to subsequent two ways,
It then first carries out step 101 and executes step 102 again.
Below still by taking above-mentioned mobile phone terminal as an example, if determining application software pair using the above-mentioned second way, i.e.,
All application software acquired in step 101 are matched two-by-two using the application software in list, obtain each application
Then there is software pair then for three application software APP1, APP2, APP3 of user 1 (IMEI1) in software pair:(APP1,
APP2)、(APP2、APP3)、(APP1、APP3)。
Step 103, for each application software pair, list is used according to acquired application software, from the M user
The concurrent users number of the middle determination application software pair and two respective users of application software of the application software centering
Number.
Wherein, determine that the concurrent users number of the application software pair is referred to from the M user from the M user
Middle determination has used the number of users of two application software of the application software centering simultaneously, obtains the application software to being total to
Deposit number of users.
It is assumed that two application software of the application software centering are respectively the first application software and the second application software, then
Two respective numbers of users of application software of the application software centering refer to having used the application software pair in M user
In the first application software number of users and M user in used the application software centering the second application software user
Number.
Fig. 2 shows the concurrent users number of application software pair, the first application software of application software centering number of users and answer
With the correlation between the number of users three of the second application software of software centering.As illustrated in fig. 2, it is assumed that 201 corresponding ellipses
Region is the number of users A of the first application software of the application software centering, and 202 corresponding elliptic regions are the application software
The number of users B of second application software of centering, then 203 corresponding intersection areas are the concurrent users number A of the application software pair
∩B。
Step 104, it for each application software pair, according to the concurrent users number of the application software pair and described answers
With two respective numbers of users of application software of software centering, be calculated the application software centering two application software it
Between correlation characterize index.
In this step, the index that can characterize the correlation between two application software of the application software centering can
There are many.
For example, three characteristic indexs being shown below in an embodiment, in practical application, can therefrom select to appoint
Meaning one or any combination indicate the correlation between the first application software and the second application software, alternatively, can also select
The index of the correlation between two application software of the application software centering can be characterized by selecting other, herein not to its into
Row limits.
1) Jie Kade (Jacard) similarity (1) can be calculated according to the following formula:
J (A, B)=(A ∩ B)/(A ∪ B) (1)
Wherein, A is the number of users of the first application software, and B is the number of users of the second application software, and A ∩ B is by the first application
The concurrent users number for the application software pair that software and the second application software are constituted, A ∪ B are that the number of users A of the first application software adds
The sum of the number of users B of second application software, then subtract the application software pair that the first application software and the second application software are constituted
Concurrent users number A ∩ B, i.e. A ∪ B=A+B- (A ∩ B).J (A, B) is Jie Kade similarity, and the value of J (A, B) shows more greatly described
Promotion degree between first application software and the second application software is bigger;Be worth it is smaller, indicate competition it is fiercer.
2) cosine (Cosine) similarity (2) can be calculated according to the following formula:
Wherein, A is the number of users of the first application software, and B is the number of users of the second application software, and A ∩ B is by the first application
The concurrent users number for the application software pair that software and the second application software are constituted, C (A, B) are cosine similarity, the value of C (A, B)
Show that the promotion degree between first application software and the second application software is bigger more greatly;Be worth it is smaller, indicate competition it is more sharp
It is strong.
The derivation process of above-mentioned formula (2) is as follows:
Assuming that the number of users of the first application software is A, corresponding vector is:Second answers
It is B with the number of users of software, corresponding vector is:Wherein, it 1 indicates to use, 0 indicates not make
With.
Then have:
3) promotion degree (Lift) coexists, (3) can be calculated according to the following formula:
L (A, B)=((A ∩ B)/B)/(A/M) (3)
Wherein, A is the number of users of the first application software, and B is the number of users of the second application software, and A ∩ B is by the first application
The concurrent users number for the application software pair that software and the second application software are constituted, M are total number of users, and L (A, B) is that promotion coexists
Degree, wherein if Lift value is greater than 1, illustrate that the probability that A occurs in B is higher than what A occurred (i.e. in M user) in deep bid
Probability, illustrates A and B is promotion relationship, it is counter know indicate competitive relation less than 1, equal to 1 indicate promote and competitive relation side
Boundary.Its value is bigger, illustrates that the promotion degree between the first application software and the second application software is bigger;It is worth smaller, expression competition
It is fiercer.
In addition, in another embodiment, it, can also be first for two application software of an application software centering
Any two in above three characteristic index or three are calculated, and as intermediate parameters, later further according to being calculated
The intermediate parameters, the correlation comprehensive score between two application software of the application software centering is calculated
Value.The specific implementation process of the process can also be there are many mode.
For example, the first:The weight for one setting of each intermediate parameters distribution being respectively calculated;According to each centre
All intermediate parameters being calculated are weighted in the weight of parameter, and two for obtaining the application software centering answer
With the correlation comprehensive grading value between software.
Second:Determine a target application software;N number of application software centering is extracted soft comprising the target application
The application software pair of part obtains R target application software pair;R is the positive integer more than or equal to 2;By the R target application
The correspondence intermediate parameters of software pair are ranked up, and obtain the sequence sequence number value of each intermediate parameters of each target application software pair;
Corresponding each target application software pair, the sequence sequence number value of each intermediate parameters of the target application software pair is added, is obtained
Correlation comprehensive grading value between two application software of the target application software centering.
Two examples that comprehensive grading value is calculated using second method are given below.
First example:Android editions synchronous correlations between assistant and some softwares.
Fig. 3 shows the calculated result of Android editions synchronous correlations between assistant and some softwares.Wherein, mesh
Mark application software is Android editions synchronous assistants, and target application software is to for each application including Android editions synchronous assistants
Software pair.In this example, while three above-mentioned intermediate parameters are calculated, and sorted to each intermediate parameters, and will
Ranking value carries out after adding up, and obtains the comprehensive grading value of each target application software pair.
Second example:Dig the correlation between money and valuables and some softwares.
Fig. 4 shows the calculated result for the correlation dug between money and valuables and some softwares.Wherein, target application software is
Money and valuables is dug, target application software is to for including each application software pair for digging money and valuables.In this example, while calculating above-mentioned three
A intermediate parameters, and sorted to each intermediate parameters, and ranking value is carried out after adding up, it is soft to obtain each target application
The comprehensive grading value of part pair.
It, can be into this method after determining the correlation characterization index between two application software of each application software centering
One step characterizes index according to the correlation between two application software of each application software centering, carries out application software to user
It presents.For example, carrying out the recommendation of respective application software to user, or launched to the advertisement that user carries out respective application software, or
It is at war with the notice etc. of application software to user.
The exemplary structure of Fig. 5 determining device of correlation between different application software a kind of in the embodiment of the present invention
Figure.Device in the present embodiment can be located in server, server cluster or cloud computing service center.The device can be used for reality
Method shown in existing Fig. 1, can be found in retouching in embodiment of the method shown in corresponding diagram 1 for undisclosed details in present apparatus embodiment
It states.As shown in figure 5, the device may include:List obtains module 501, software to determining module 502, number of users determining module 503
With index computing module 504.
Wherein, list obtains module 501 and uses list for obtaining the application software of M user;M is more than or equal to 2
Positive integer.It is corresponding with method shown in Fig. 1, what the intelligent terminal that list acquisition module 501 can be used for receiving M user uploaded
Application software uses list;Alternatively, the application software for obtaining M user from each application software server uses list.
Software is to determining module 502 for determining N number of application software pair;N is the positive integer more than or equal to 1.With Fig. 1 institute
Show that method is corresponding, software it is soft can to predefine application to be measured according to user demand or other strategies to determining module 502
Part pair.It is carried out alternatively, list can also be obtained all application software acquired in module 501 using the application software in list
It matches two-by-two, obtains each application software pair.Alternatively, can also be the application software according to acquired in list acquisition module 501
Using list, the N number of application software pair for meeting setting demand is determined.
In the present embodiment, if if software predefines application software pair to be measured to determining module 502, list is obtained
Module 501 and software is not to that can have data flow between determining module 502, at this point, list acquisition 501 need of module will be acquired
Application software be supplied to number of users determining module 503 using list.If software uses subsequent two kinds of sides to determining module 502
If formula determines application software pair, then list obtains module 501 and software to there are data flows between determining module 502, at this point,
Acquired application software is supplied to software using list and determined to determining module 502 and number of users by list acquisition module 501
Module 503.
Number of users determining module 503 is used to be directed to each application software pair, uses list according to acquired application software,
The concurrent users number of the application software pair and two applications of the application software centering are determined from the M user
The respective number of users of software.Wherein, determine that the concurrent users number of the application software pair is referred to from institute from the M user
The number of users that two application software of the application software centering are determined while used in M user is stated, the application is obtained
The concurrent users number of software pair.It is assumed that two application software of the application software centering are respectively the first application software and second
Application software, then two respective numbers of users of application software of the application software centering refer to having used institute in M user
Stating in the number of users and M user of the first application software of application software centering has used the second of the application software centering to answer
With the number of users of software.
Index computing module 504 be used for be directed to each application software pair, according to the concurrent users number of the application software pair,
And two respective numbers of users of application software of the application software centering, two of the application software centering are calculated
Correlation between application software characterizes index.
Corresponding with method shown in Fig. 1, in one embodiment, two applications that index computing module 504 calculates are soft
Correlation between part characterizes index:Jie Kade similarity, cosine similarity between two application software and altogether
Deposit any one in promotion degree or any combination.Wherein, the Jie Kade similarity can be calculated according to above-mentioned formula (1)
It arrives, the cosine similarity can be calculated according to above-mentioned formula (2), and the promotion degree that coexists can be according to above-mentioned formula
(3) it is calculated.
Corresponding with method shown in Fig. 1, in another embodiment, index computing module 504 may include:Intermediate parameters
Computing module 5041 and comprehensive score determining module 5042.
Wherein, intermediate parameters computing module 5041 is used to calculate between two application software of the application software centering
Jie Kade similarity, cosine similarity and any two in promotion degree or three coexist, obtains two or three intermediate parameters.
Wherein, the Jie Kade similarity can be calculated according to above-mentioned formula (1), and the cosine similarity can be according to above-mentioned public affairs
Formula (2) is calculated, and the promotion degree that coexists can be calculated according to above-mentioned formula (3).
Comprehensive score determining module 5042 is used to that it is soft that the application to be calculated according to the intermediate parameters being calculated
Correlation comprehensive grading value between two application software of part centering.
When specific implementation, two of them can be set forth below there are many way of realization in comprehensive score determining module 5042.
Fig. 6 shows the structural schematic diagram of comprehensive score determining module in an example of the invention.As shown in fig. 6, the example
In son, the comprehensive score determining module 5042 can further comprise:Weight distribution module 601 and score value computing module 602.
Weight distribution module 601 is for being respectively that each intermediate parameters being calculated distribute a weight set.
Score value computing module 602 is used for the weight according to each intermediate parameters, to all intermediate parameters being calculated
It is weighted, obtains the correlation comprehensive grading value between two application software of the application software centering.
Fig. 7 shows the structural schematic diagram of comprehensive score determining module in another example of the invention.As shown in fig. 7, should
In example, the comprehensive score determining module 5042 can further comprise:Target software determining module 701, target software are to true
Cover half block 702, sorting module 703 and score value computing module 704.
Wherein, target software determining module 701 is for determining a target application software.
Target software is used to extract from N number of application software centering soft comprising the target application to determining module 702
The application software pair of part obtains R target application software pair.Wherein, R is the positive integer more than or equal to 2.
Sorting module 703 obtains each for being ranked up the correspondence intermediate parameters of the R target application software pair
The sequence sequence number value of each intermediate parameters of target application software pair.
Score value computing module 704 is for corresponding to each target application software pair, by each of the target application software pair
The sequence sequence number value of intermediate parameters is added, and obtains the correlation between two application software of the target application software centering
Comprehensive grading value.
Fig. 8 is a kind of structural schematic diagram of server in the embodiment of the present invention.The server can be used for realizing side shown in Fig. 1
Method and Fig. 5 shown device.As shown in figure 8, the server may include:Processor 801, non-volatile computer readable memory
802, display unit 803 and network communication interface 804.These components are communicated by bus 805.
In the present embodiment, multiple program modules are stored in memory 802, such as:Operating system 806, network communication mould
Block 807 and application program 808.
Processor 801 can read various module (not shown)s included by the application program 808 in memory 802
To execute the various function application and data processing of user terminal.Processor 801 in the present embodiment can be one, can also
Think it is multiple, can be CPU, processing unit/module, ASIC, logic module or programmable gate array etc..
Wherein, operating system 806 includes but is not limited to:Android operation system, Windows operating system, apple iOS
Operating system, apple Mac OS operating system etc..
Application program 808 may include each functional module in Fig. 5 shown device, and it is executable to form corresponding computer
Instruction set 809 and corresponding metadata and heuritic approach 810.These set of computer-executable instructions can be by the processor
801 execute and complete the function of method shown in Fig. 1 or Fig. 5 shown device.
In the present embodiment, network communication interface 804 is matched with network communication module 807 completes the various networks of server
The transmitting-receiving of signal uses transmission of list and network data exchange etc. including the application software between intelligent terminal.
Display unit 803 has a display panel, for completing the input and display of relevant information, including presentation using soft
Correlation etc. between part.
The present invention also provides a kind of storage medium, such as non-volatile computer readable storage medium storing program for executing, wherein being stored with number
According to processing routine, which is used to execute any embodiment of the above method of the present invention.
The exemplary structure of Fig. 9 determination system of correlation between different application software a kind of in the embodiment of the present invention
Figure.As shown in figure 9, the system includes:M intelligent terminal 901 and phase such as between Fig. 5 or different application software shown in fig. 8
The determining device 902 of mutual relation.
Wherein, each intelligent terminal 901 in the M intelligent terminal 901 is for making the application software of own user
Described device 902 is supplied to list.M is the positive integer more than or equal to 2.Intelligent terminal 901 can be mobile phone, plate electricity
Brain, pocket computer on knee and desktop computer etc..
It can be connected by wireless network or cable network between intelligent terminal 901 and described device 902.
The application software that described device 902 is used to obtain M user uses list;M is the positive integer more than or equal to 2;
Determine N number of application software pair;N is the positive integer more than or equal to 1;For each application software pair, according to acquired application
Software uses list, the concurrent users number of the determining application software pair and the application software pair from the M user
In two respective numbers of users of application software;According to the concurrent users number of the application software pair and the application software
Two respective numbers of users of application software of centering, are calculated the phase between two application software of the application software centering
Mutual relation characteristic index.Device 902 can be a server, or the server cluster consisted of several servers, or
Person is cloud computing service center etc..
In addition, can also provide a kind of application software in the application presents system, which may include that above-mentioned difference is answered
Device is presented with the determining device and one of correlation between software.
Wherein, presenting device can respectively answer according to determined by the determining device of correlation between the different application software
Index is characterized with the correlation between two application software of software centering, application software is carried out to user and presents.For example, to
User carries out the recommendation of respective application software, or launches to the advertisement that user carries out respective application software, or competing to user's progress
Strive the notice etc. of application software.
In the embodiment of the present invention, the application software by obtaining multiple users uses list, and according to acquired application
Software matrix determines multiple application software pair;For each application software pair, list is used according to acquired application software, from
Two applications of concurrent users number and the application software centering that the application software pair is determined in the M user are soft
The respective number of users of part, and then the correlation characterization being calculated between two application software of the application software centering refers to
Mark.As it can be seen that the characteristic index based on objective data, accurately has found the correlation between application software, and its
It is presented in a manner of numerical value, realizes the quantization to correlation between two application software, so that not only may determine that two
Correlation between application software is promotion relationship or competitive relation, and can quantify its degree for promoting or competing.
It is possible to further using between two application software Jie Kade similarity, cosine similarity and promotion coexists
Any one or any combination in degree characterize the correlation between two application software, and value is bigger, indicates to promote journey
Degree is bigger, and value is smaller, indicates that competition is fiercer.
Further, it is also possible to according among above three index any two or three, according to preset weight or sequence
Value, is calculated a comprehensive grading value, so as to the correlation between two application software of more intuitive determination.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (17)
1. a kind of determination method of correlation between different application software, which is characterized in that including:
The application software for obtaining M user uses list;M is the positive integer more than or equal to 2;
Determine N number of application software pair;N is the positive integer more than or equal to 1;
For each application software pair, list is used according to acquired application software, is answered described in determination from the M user
With the concurrent users number of software pair and two respective numbers of users of application software of the application software centering;
For each application software pair, according to the concurrent users number of the application software pair and the application software centering
Two respective numbers of users of application software, are calculated the correlation between two application software of the application software centering
Characteristic index;
The concurrent users number that the application software pair is determined from the M user includes:
The number of users that two application software of the application software centering are determined while used from the M user, obtains
The concurrent users number of the application software pair.
2. the method according to claim 1, wherein the N number of application software of the determination is to including:
Predefine N number of application software pair to be measured;Alternatively,
Acquired all application software are matched two-by-two using the application software in list, obtain N number of application software pair;
Alternatively,
List is used according to acquired application software, determines the N number of application software pair for meeting setting demand.
3. the method according to claim 1, wherein the correlation characterization between described two application software refers to
Mark includes:Jie Kade similarity, cosine similarity between two application software and any one in promotion degree coexists or appoints
Meaning combination;Wherein,
The Jie Kade similarity is calculated according to the following formula:
J (A, B)=(A ∩ B)/(A ∪ B)
The cosine similarity is calculated according to the following formula:
The promotion degree that coexists is calculated according to the following formula:
L (A, B)=((A ∩ B) B)/(AM)
Wherein, A is the number of users of the first application software of the application software centering, and B is the second of the application software centering
The number of users of application software, M are total number of users, and A ∩ B is the concurrent users number of the application software pair, and A ∪ B is A+B- (A ∩
B), J (A, B) is Jie Kade similarity, and C (A, B) is cosine similarity, and L (A, B) is that promotion degree coexists.
4. the method according to claim 1, wherein two applications that the application software centering is calculated are soft
Correlation between part characterizes index:
It calculates the Jie Kade similarity between two application software of the application software centering, cosine similarity and promotion coexists
Any two in degree or three, obtain two or three intermediate parameters;
According to the intermediate parameters being calculated, the phase between two application software of the application software centering is calculated
Mutual relation comprehensive grading value;
Wherein, the Jie Kade similarity is calculated according to the following formula:
J (A, B)=(A ∩ B)/(A ∪ B)
The cosine similarity is calculated according to the following formula:
The promotion degree that coexists is calculated according to the following formula:
L (A, B)=((A ∩ B)/B)/(AM)
Wherein, A is the number of users of the first application software of the application software centering, and B is the second of the application software centering
The number of users of application software, M are total number of users, and A ∩ B is the concurrent users number of the application software pair, and A ∪ B is A+B- (A ∩
B), J (A, B) is Jie Kade similarity, and C (A, B) is cosine similarity, and L (A, B) is that promotion degree coexists.
5. according to the method described in claim 4, it is characterized in that, the intermediate parameters that the basis is calculated, calculate
The correlation comprehensive grading value obtained between two application software of the application software centering includes:
The weight for one setting of each intermediate parameters distribution being respectively calculated;
According to the weight of each intermediate parameters, all intermediate parameters being calculated are weighted, the application is obtained
Correlation comprehensive grading value between two application software of software centering.
6. according to the method described in claim 4, it is characterized in that, the intermediate parameters that the basis is calculated, calculate
The correlation comprehensive grading value obtained between two application software of the application software centering includes:
Determine a target application software;
The application software pair comprising the target application software is extracted from N number of application software centering, obtains R target application
Software pair;R is the positive integer more than or equal to 2;
The correspondence intermediate parameters of the R target application software pair are ranked up, each of each target application software pair is obtained
The sequence sequence number value of intermediate parameters;
Corresponding each target application software pair, the sequence sequence number value of each intermediate parameters of the target application software pair is added,
Obtain the correlation comprehensive grading value between two application software of the target application software centering.
7. the method according to claim 1, wherein the application software for obtaining M user uses list packet
It includes:
The application software for receiving the intelligent terminal upload of M user uses list;Alternatively,
The application software that M user is obtained from each application software server uses list.
8. method according to any one of claim 1 to 7, which is characterized in that this method further comprises:According to respectively answering
Index is characterized with the correlation between two application software of software centering, application software is carried out to user and presents.
9. the determining device of correlation between a kind of different application software, which is characterized in that including:
List obtains module, and the application software for obtaining M user uses list;M is the positive integer more than or equal to 2;
Software is to determining module, for determining N number of application software pair;N is the positive integer more than or equal to 1;
Number of users determining module uses list according to acquired application software, from described for being directed to each application software pair
Two application software of concurrent users number and the application software centering that the application software pair is determined in M user are each
From number of users;
Index computing module, for being directed to each application software pair, according to concurrent users number, the Yi Jisuo of the application software pair
Two respective numbers of users of application software for stating application software centering, two applications that the application software centering is calculated are soft
Correlation between part characterizes index;
Wherein, the number of users determining module is used for, and is determined from the M user while having been used the application software centering
Two application software number of users, obtain the concurrent users number of the application software pair.
10. device according to claim 9, which is characterized in that the software is to be measured for predefining to determining module
N number of application software pair;Alternatively, the list, which is obtained all application software that module obtains, uses the application software in list
It is matched two-by-two, obtains N number of application software pair;Alternatively, obtaining the application software that module obtains according to the list uses column
Table determines the N number of application software pair for meeting setting demand.
11. device according to claim 9, which is characterized in that two application software that the index computing module calculates
Between correlation characterization index include:Jie Kade similarity, cosine similarity between two application software and coexisting mentions
Any one in liter degree or any combination;Wherein,
The Jie Kade similarity is calculated according to the following formula:
J (A, B)=(A ∩ B)/(A ∪ B)
The cosine similarity is calculated according to the following formula:
The promotion degree that coexists is calculated according to the following formula:
L (A, B)=((A ∩ B)/B)/(AM)
Wherein, A is the number of users of the first application software of the application software centering, and B is the second of the application software centering
The number of users of application software, M are total number of users, and A ∩ B is the concurrent users number of the application software pair, and A ∪ B is A+B- (A ∩
B), J (A, B) is Jie Kade similarity, and C (A, B) is cosine similarity, and L (A, B) is that promotion degree coexists.
12. device according to claim 9, which is characterized in that the index computing module includes:
Intermediate parameters computing module, the Jie Kade between two application software for calculating the application software centering are similar
Cosine similarity and any two in promotion degree or three coexist in degree, obtain two or three intermediate parameters;
Comprehensive score determining module, for the application software centering to be calculated according to the intermediate parameters being calculated
Two application software between correlation comprehensive grading value;
Wherein, the Jie Kade similarity is calculated according to the following formula:
J (A, B)=(A ∩ B)/(A ∪ B)
The cosine similarity is calculated according to the following formula:
The promotion degree that coexists is calculated according to the following formula:
L (A, B)=((A ∩ B)/B)/(AM)
Wherein, A is the number of users of the first application software of the application software centering, and B is the second of the application software centering
The number of users of application software, M are total number of users, and A ∩ B is the concurrent users number of the application software pair, and A ∪ B is A+B- (A ∩
B), J (A, B) is Jie Kade similarity, and C (A, B) is cosine similarity, and L (A, B) is that promotion degree coexists.
13. device according to claim 12, which is characterized in that the comprehensive score determining module includes:
Weight distribution module, for being respectively the weight for one setting of each intermediate parameters distribution being calculated;
Score value computing module adds all intermediate parameters being calculated for the weight according to each intermediate parameters
Power calculates, and obtains the correlation comprehensive grading value between two application software of the application software centering.
14. device according to claim 12, which is characterized in that the comprehensive score determining module includes:
Target software determining module, for determining a target application software;
Target software is to determining module, for extracting answering comprising the target application software from N number of application software centering
With software pair, R target application software pair is obtained;R is the positive integer more than or equal to 2;
Sorting module obtains each target and answers for being ranked up the correspondence intermediate parameters of the R target application software pair
With the sequence sequence number value of each intermediate parameters of software pair;
Score value computing module, for corresponding to each target application software pair, by each intermediate ginseng of the target application software pair
Several sequence sequence number values is added, and the correlation synthesis obtained between two application software of the target application software centering is commented
Score value.
15. device according to claim 9, which is characterized in that the list obtains the intelligence that module is used to receive M user
The application software that energy terminal uploads uses list;Alternatively, obtaining the application software of M user from each application software server makes
Use list.
16. the determination system of correlation between a kind of different application software, which is characterized in that including:M intelligent terminal and such as
The determining device of correlation between a kind of different application software described in any one of claim 9 to 15;Wherein,
Each intelligent terminal in the M intelligent terminal is used to for the application software of own user being supplied to using list described
Device;M is the positive integer more than or equal to 2.
17. a kind of application software presents system, which is characterized in that including:Present device and such as any one of claim 9 to 15
The determining device of correlation between a kind of different application software;Wherein,
The device of presenting according to determined by the determining device of correlation between the different application software for respectively applying
Correlation between two application software of software centering characterizes index, carries out application software to user and presents.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610559788.9A CN106682056B (en) | 2016-07-15 | 2016-07-15 | The determination method, apparatus and system of correlation between different application software |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610559788.9A CN106682056B (en) | 2016-07-15 | 2016-07-15 | The determination method, apparatus and system of correlation between different application software |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106682056A CN106682056A (en) | 2017-05-17 |
CN106682056B true CN106682056B (en) | 2018-11-20 |
Family
ID=58839837
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610559788.9A Active CN106682056B (en) | 2016-07-15 | 2016-07-15 | The determination method, apparatus and system of correlation between different application software |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106682056B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109213513B (en) * | 2017-06-30 | 2021-07-27 | 腾讯科技(深圳)有限公司 | Method and device for determining share ratio of software and computer readable storage medium |
CN108355349A (en) * | 2018-03-14 | 2018-08-03 | 张伟东 | Games system |
CN110163460B (en) * | 2018-03-30 | 2023-09-19 | 腾讯科技(深圳)有限公司 | Method and equipment for determining application score |
CN111949860B (en) * | 2019-05-15 | 2022-02-08 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating a relevance determination model |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104750798A (en) * | 2015-03-19 | 2015-07-01 | 腾讯科技(深圳)有限公司 | Application program recommendation method and device |
CN104778178A (en) * | 2014-01-13 | 2015-07-15 | 腾讯科技(深圳)有限公司 | Application classification method, application classification device and service server |
US9213729B2 (en) * | 2012-01-04 | 2015-12-15 | Trustgo Mobile, Inc. | Application recommendation system |
CN105608180A (en) * | 2015-12-22 | 2016-05-25 | 北京奇虎科技有限公司 | Application recommendation method and system |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102541590B (en) * | 2011-12-14 | 2015-09-30 | 北京奇虎科技有限公司 | A kind of software recommendation method and commending system |
EP2960764A1 (en) * | 2014-06-27 | 2015-12-30 | Orange | Method for selecting an entry for an application using a graphical user interface |
CN107644079A (en) * | 2015-05-22 | 2018-01-30 | 广东欧珀移动通信有限公司 | Method and device and related media production are recommended in one kind application |
-
2016
- 2016-07-15 CN CN201610559788.9A patent/CN106682056B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9213729B2 (en) * | 2012-01-04 | 2015-12-15 | Trustgo Mobile, Inc. | Application recommendation system |
CN104778178A (en) * | 2014-01-13 | 2015-07-15 | 腾讯科技(深圳)有限公司 | Application classification method, application classification device and service server |
CN104750798A (en) * | 2015-03-19 | 2015-07-01 | 腾讯科技(深圳)有限公司 | Application program recommendation method and device |
CN105608180A (en) * | 2015-12-22 | 2016-05-25 | 北京奇虎科技有限公司 | Application recommendation method and system |
Non-Patent Citations (1)
Title |
---|
移动终端的APP应用组织研究综述;赵怡然;《信息通信》;20151231(第1期);第230-231页 * |
Also Published As
Publication number | Publication date |
---|---|
CN106682056A (en) | 2017-05-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106682056B (en) | The determination method, apparatus and system of correlation between different application software | |
CN105306495B (en) | user identification method and device | |
CN106487747B (en) | User identification method, system, device and processing method, device | |
CN104808892B (en) | A kind of application icon sort method, device, system and relevant device | |
CN110782318A (en) | Marketing method and device based on audio interaction and storage medium | |
CN110785970B (en) | Techniques to automate robotic creation of web pages | |
CN109471978B (en) | Electronic resource recommendation method and device | |
CN110347501A (en) | A kind of service testing method, device, storage medium and electronic equipment | |
CN103838884B (en) | Messaging device and information processing method | |
CN104462051B (en) | Segmenting method and device | |
CN113412608B (en) | Content pushing method and device, server and storage medium | |
CN105531701A (en) | Personalized trending image search suggestion | |
CN107729492A (en) | Exercise pushing method and system and terminal equipment | |
CN110414581B (en) | Picture detection method and device, storage medium and electronic device | |
CN109299742A (en) | Method, apparatus, equipment and the storage medium of automatic discovery unknown network stream | |
CN108776692A (en) | Method and apparatus for handling information | |
CN104915399A (en) | Recommended data processing method based on news headline and recommended data processing method system based on news headline | |
CN109189960A (en) | Information acquisition method, information acquisition device and computer storage medium | |
CN108319888A (en) | The recognition methods of video type and device, terminal | |
CN110020156A (en) | Information recommendation method, front end implementation method, device, equipment and storage medium | |
Ye et al. | Identifying influential individuals on large-scale social networks: A community based approach | |
CN109495513A (en) | Unsupervised encryption malicious traffic stream detection method, device, equipment and medium | |
CN109002434A (en) | Customer service question and answer matching method, server and storage medium | |
CN108537291A (en) | A kind of sample classification method and apparatus | |
CN107729931A (en) | Picture methods of marking and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |