CN109213513A

CN109213513A - The determination method, apparatus and computer readable storage medium of software share accounting

Info

Publication number: CN109213513A
Application number: CN201710521670.1A
Authority: CN
Inventors: 谢毅; 胡荣杰
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2017-06-30
Filing date: 2017-06-30
Publication date: 2019-01-15
Anticipated expiration: 2037-06-30
Also published as: CN109213513B

Abstract

The invention discloses a kind of determination method, apparatus of software share accounting and computer readable storage mediums, belong to technical field of data processing.This method comprises: determining designated software of the installation probability to target software with inhibition or castering action from multiple designated softwares；Determine the joint probability between target software and identified designated software；Count the first condition probability of the corresponding terminal installation targets software at least two dimensions under conditions of installing identified designated software and target software；Based on the joint probability and first condition probability between target software and identified designated software, identified joint probability between designated software, target software and at least two dimensions is determined；Based on identified joint probability, the dimensional information of multiple sample datas of storage and multiple sample datas, the software share accounting of target software is obtained.Due to being not necessarily to the veteran data statistics personnel manual operation of industry background, this improves determine efficiency.

Description

The determination method, apparatus and computer readable storage medium of software share accounting

Technical field

The present invention relates to technical field of data processing, in particular to the determination method, apparatus of a kind of software share accounting and Computer readable storage medium.

Background technique

Currently, operator provides rich and varied software, for the ease of understanding the use feelings of software in the market in depth Condition and formulation product strategy, operator are often desirable to know that software share accounting, software share accounting refer to that certain software exists The share accounting being mounted in market.In order to determine the software share accounting, some data statistics quotient provide a data and adopt Collect SDK (Software Development Kit, Software Development Kit), data acquisition SDK can have in some soft In part, for example, being present in using in house keeper, in this way, can pass through the data when being equipped with the software in terminal and acquire SDK Obtain the software matrix of simultaneously the installed software of reporting terminal itself, the software reported in order to data statistics quotient according to each terminal List determines the software share accounting of each software by the way of sampling.

In practical application scene, terminal models, user's gender, age of user, Promotion Strategy of data acquisition SDK etc. are tieed up Degree will affect the accuracy of identified software share accounting.For example, by taking terminal models as an example, if the terminal of certain type is not Support installation software A, in this way, when counting the software share accounting of software A, if most of software matrix of sampling is It is reported by the terminal of the type, is then easy to cause counted software share accounting that there is certain skewed popularity.For this purpose, in phase In the technology of pass, generally requires the data statistics personnel with enough industry background experiences and analyzes and correct, it is generally the case that Data statistics personnel can rule of thumb know which dimension results in software share accounting with skewed popularity, and be directed to the dimension Degree re-starts sampling, with the software share accounting of statistical software again, to reach modified effect.

During realizing the application, the inventor finds that the existing technology has at least the following problems: due in related skill The veteran data statistics personnel of industry background are needed to carry out manual analysis and amendment in art, in this way, software share is caused to account for The determination efficiency of ratio is lower.

Summary of the invention

Determination efficiency in order to solve the problems, such as the software share accounting of the relevant technologies is lower, and the embodiment of the invention provides one The determination method and device of kind software share accounting.The technical solution is as follows:

In a first aspect, providing a kind of determination method of software share accounting, which comprises

Determine that the installation probability to target software has the designated software of inhibition or castering action from multiple designated softwares, Each designated software refers to that the installation probability to preset quantity other software all has the software of inhibition or castering action, described Target software is the software of software share accounting to be counted；

Determine the joint probability between the target software and identified designated software；

It is corresponding at least two dimensions under conditions of statistics designated software determined by installing and the target software Terminal installs the first condition probability of the target software, and each dimension at least two dimension influences the target The software share accounting of software；

Based between the target software and identified designated software joint probability and the first condition probability, really Joint probability determined by fixed between designated software, the target software and at least two dimension；

Based between identified designated software, the target software and at least two dimension joint probability, deposit Multiple sample datas of storage and the dimensional information of the multiple sample data, obtain the software share accounting of the target software.

Second aspect provides a kind of determination method of software share accounting, which comprises

The marginal probability of corresponding terminal installation targets software in target dimension is obtained, the target software is to be counted soft The software of part share accounting, the target dimension are at least two dimensions of software share accounting for influence the target software Dimension with proper boundary probability；

Statistics is described at least two under conditions of terminal corresponding in the target dimension has installed the target software Corresponding terminal installs the third condition probability of the target software in other dimensions in a dimension；

Based on the marginal probability and the third condition probability, determine that the joint between at least two dimension is general Rate；

Dimensional information based on the joint probability, multiple sample datas of storage and the multiple sample data, obtains The software share accounting of the target software.

The third aspect, provides a kind of determining device of software share accounting, and described device includes:

First determining module inhibits or mentions for determining to have the installation probability of target software from multiple designated softwares The designated software of effect is risen, each designated software, which refers to, to be all had inhibition to the installation probability of preset quantity other software or mention The software of effect is risen, the target software is the software of software share accounting to be counted；

Second determining module, for determining the joint probability between the target software and identified designated software；

First statistical module, under conditions of for counting the designated software determined by installing and the target software at least Corresponding terminal installs the first condition probability of the target software, each dimension at least two dimension in two dimensions Degree influences the software share accounting of the target software；

Third determining module, for based between the target software and identified designated software joint probability and institute First condition probability is stated, determines identified joint between designated software, the target software and at least two dimension Probability；

4th determining module, for based on identified designated software, the target software and at least two dimension Between joint probability, the dimensional information of multiple sample datas of storage and the multiple sample data, it is soft to obtain the target The software share accounting of part.

Fourth aspect, provides a kind of determining device of software share accounting, and described device includes:

Module is obtained, for obtaining the marginal probability of corresponding terminal installation targets software in target dimension, the target Software is the software of software share accounting to be counted, and the target dimension is to influence the software share accounting of the target software With the dimension of proper boundary probability at least two dimensions；

First statistical module has installed the target software for counting the corresponding terminal in the target dimension Under the conditions of corresponding terminal installs the third condition probability of the target software in other dimensions at least two dimension；

First determining module determines described at least two for being based on the marginal probability and the third condition probability Joint probability between dimension；

Second statistical module, for based on the joint probability, storage multiple sample datas and the multiple sample number According to dimensional information, obtain the software share accounting of the target software.

5th aspect, provides a kind of terminal, the terminal includes processor and memory, is stored in the memory At least one instruction, at least one section of program, code set or instruction set, described instruction, described program, the code set or the finger Collection is enabled to be loaded as the processor and executed to realize that software share accounting is really described in above-mentioned first aspect or second aspect Determine method.

6th aspect, provides a kind of computer readable storage medium, at least one finger is stored in the storage medium It enables, at least one section of program, code set or instruction set, described instruction, described program, the code set or described instruction collection are by handling Device loads and executes the determination method to realize software share accounting described in above-mentioned first aspect or second aspect.

Technical solution provided in an embodiment of the present invention has the benefit that and determines from multiple designated softwares to target The installation probability of software has the designated software of inhibition or castering action, and determines between the target software and the designated software Joint probability.Later, statistics is corresponding at least two dimensions under conditions of installing identified designated software and target software Terminal installation targets software first condition probability, it is general based on combining between the target software and identified designated software Rate and the first condition probability determine identified joint between designated software, the target software and at least two dimension Probability determines software share accounting with the dimensional information based on the joint probability and multiple sample datas and multiple sample datas. In the process, it is contemplated that the conditional relationship between designated software and at least two dimensions and the Joint Distribution to the target software It influences, ensure that the accuracy of determining software share accounting, also, due to being not necessarily to during determining software share accounting The veteran data statistics personnel of industry background are manually participated in, and this improves determine efficiency.

Detailed description of the invention

To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment 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.

Fig. 1 is a kind of schematic diagram of application scenarios provided in an embodiment of the present invention；

Fig. 2A is a kind of flow chart of the determination method of software share accounting provided in an embodiment of the present invention；

Fig. 2 B is a kind of schematic diagram of correction result shown according to an exemplary embodiment；

Fig. 2 C is a kind of schematic diagram of the correction result shown according to another exemplary embodiment；

Fig. 3 A is the flow chart of the determination method of another software share accounting provided in an embodiment of the present invention；

Fig. 3 B is a kind of flow diagram of the determination method of software share accounting involved in Fig. 3 A embodiment；

Fig. 3 C is that a kind of effect of software share accounting determined based on terminal models shown in the embodiment of the present invention is illustrated Figure；

Fig. 4 A is a kind of structural schematic diagram of the determining device of software share accounting provided in an embodiment of the present invention；

Fig. 4 B is the structural schematic diagram of the determining device of another software share accounting provided in an embodiment of the present invention；

Fig. 5 A is the structural schematic diagram of the determining device of another software share accounting provided in an embodiment of the present invention；

Fig. 5 B is the structural schematic diagram of the determining device of another software share accounting provided in an embodiment of the present invention；

Fig. 6 is a kind of server architecture schematic diagram of the determining device of software share accounting provided in an embodiment of the present invention.

Specific embodiment

To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention Formula is described in further detail.

Before to the embodiment of the present invention carrying out that explanation is explained in detail, first to the present embodiments relate to noun, answer For a brief explanation with scene and implementation environment.

Firstly, to the present embodiments relate to noun be illustrated.

Software share accounting:Refer to the share accounting that software is mounted in the market.Operator accounts for according to the software share Situation about being mounted in the market than that can know software, that is to say, operator can know soft according to the software share accounting The service condition of part in the market.

KOL:(Key Opinion Development, leader of opinion), it is to be understood that being a kind of software, the software There is inhibition or castering action to the installation probability of other softwares, for example, the software is that application is precious, usually due to the application treasured The software of such as QQ etc can be promoted, therefore, the installation probability of the precious software to such as QQ etc of the application has certain mention Rise effect.

Gibbs sampling:It is to be used in the special karr theory of (Markov Chain Monte Carlo, MCMC) Markov chain illiteracy To obtain a series of the distribution of specified multidimensional probability is approximately equal to (for example, referring between multiple dimensions in embodiments of the present invention Joint probability), with the algorithm of sample drawn data.

Promotion degree:It is commonly used for whether one software of instruction has promotion or suppression to the installation probability of another software Production is used.In practical applications, if installing the second software under conditions of installing the first software, when first software and this When promotion degree between two softwares is greater than 1, illustrate that first software has castering action to the installation probability of second software, when When promotion degree between first software and second software is less than 1, illustrate that first software is general to the installation of second software Rate is inhibited.

Secondly, the application scenarios to the embodiment of the present invention are illustrated.

In practical application scene, when counting the software share accounting of target software, not only terminal models, Yong Huxing Not, the dimensions such as Promotion Strategy of age of user, data acquisition SDK will affect the accuracy of the determination of software share accounting, some Fraternal software product or competition software product also will affect the accuracy of the determination of software share accounting.For example, referring to FIG. 1, A software if it exists, the software have castering action to the installation probability of the target software of current statistic software share accounting, Then the software is higher by the software share accounting that will lead to the target software, whereas if there are a software, the software is to working as The installation probability of the target software of preceding statistical software share accounting is inhibited, then the software will will lead to the target software Software share accounting it is less than normal.Under normal conditions, such software is referred to as KOL.

Therefore, it when there is the installation probability to target software has the other software of promotion or inhibiting effect, is using During the methods of sampling counts software share accounting, it is also necessary to consider the target software and can to influence the target soft Joint Distribution situation between the other software of part.

For this purpose, the software share accounting is really the embodiment of the invention provides a kind of determination method of software share accounting Determining method not only and can solve to only exist at least two dimensions has the problem of deviation influences to software share accounting, can also solve While certainly there are at least two dimensions there is deviation to influence software share accounting, the above-mentioned software for belonging to KOL is soft to target The software share accounting of part also has the problem of deviation influences.

Finally, the implementation environment to the embodiment of the present invention is illustrated.

The determination method of software share accounting provided in an embodiment of the present invention can by server as executing subject Lai real It is existing.In the concrete realization, which can be a server, or the server set being made of multiple servers Group, it is not limited in the embodiment of the present invention.

Wherein, which can receive the sample data reported from each data source, wherein the sample data is specific It can be software matrix.Multiple data source may include the terminal of such as mobile phone, tablet computer etc, can pacify in the terminal Equipped with the software for carrying data statistics SDK, for example, the software can be for using treasured, sofeware management etc..In this way, terminal can lead to The software matrix of the installed software of the SDK acquisition terminal in the software is crossed, and the software matrix is reported to the server.It Afterwards, which can be based on the received all sample datas of institute, the software share accounting provided through the embodiment of the present invention It determines method, determines that the software share accounting of software, specific implementation process may refer to real shown in following Fig. 2A and Fig. 3 A Apply example.

As it was noted above, when counting the software share accounting of target software, there are two kinds of feelings in practical application scene Condition: the first situation, influence software share accounting includes above-mentioned at least two dimension, and further includes to the target software Installation probability has the other software centainly influenced；Second situation, influence software share accounting only includes such as above-mentioned end At least two dimensions of terminal type, user's gender, age of user etc.It in order to facilitate understanding and reads, next will lead to respectively It crosses two embodiments of following Fig. 2A and Fig. 3 A and is directed to above-mentioned two situations, the specific implementation to the determination method of software share accounting Process is described in detail.

It firstly, including above-mentioned at least two dimension with influence software share accounting, and further include to the target software Installation probability there are the other software that centainly influence for be illustrated, please refer to Fig. 2A, which is the embodiment of the present invention A kind of flow chart of the determination method of the software share accounting provided, this method are applied in server.This method comprises:

Step 201: determining that the installation probability to target software has from multiple designated softwares and inhibit or castering action Designated software, each designated software refer to that the installation probability to preset quantity other software all has inhibition or castering action Software, the target software are the software of software share accounting to be counted.

Wherein, determined from multiple designated softwares the installation probability to target software have inhibit or castering action it is specified The specific implementation of software may include: the promotion in determining multiple designated software between each designated software and the target software Degree, if there is the promotions degree between the target software greater than 1 or the designated software less than 1, it is determined that the designated software is There is the designated software of inhibition or castering action to the installation probability of the target software.

Wherein it is determined that the specific reality of the promotion degree in multiple designated software between each designated software and the target software It now include: the marginal probability of the designated software to be obtained for each designated software in the designated software, and statistics is being installed The conditional probability that the target software is installed under conditions of the designated software, by the conditional probability that is counted divided by acquisition should The marginal probability of designated software is to get the promotion degree arrived between the designated software and the target software.

In one possible implementation, the marginal probability of each designated software is stored in the server, in reality In realization, the marginal probability of each designated software stored after being acquired by the publication channel of such as Ministry of Industry and Information etc to In the server, alternatively, the marginal probability of each designated software can also be obtained by carrying out data exchange between third party After store into server, it is not limited in the embodiment of the present invention.

In addition, in practical implementations, the sample data which can report according to multiple data sources, statistics is being installed The conditional probability that the target software is installed under conditions of the designated software, that is to say, in the multiple data source of server statistics In the sample data of report, sample size S1 corresponding to the target software is installed under conditions of being mounted with the designated software, with And count in the sample data that multiple data source reports, sample size S2 corresponding to the target software is installed.Later, by this Sample size S1 is divided by sample size S2 to get to installing the target software under conditions of being mounted with the designated software Conditional probability.

Further, determining that there is inhibition or castering action to the installation probability of target software from multiple designated softwares Designated software before, it is also necessary to determine multiple designated software, in the concrete realization, determine the realization of multiple designated software In the process may include following 2011-2014:

2011: determining the first software ratio shared in multiple sample data, which is multiple sample Any software in the corresponding multiple softwares of data.

In practical implementations, the first software of any of multiple softwares corresponding for multiple sample data, needs First software share shared in multiple sample data is determined, the reason is that, if first software is in multiple samples Shared share is smaller in data, then illustrates that the influence power of first software is also smaller, can ignore, but if this is first soft Part share shared in multiple sample data is larger, illustrates that first software has large effect power, at this time cannot Ignore the first software bring influence.It is thus necessary to determine that the ratio that first software is shared in multiple sample data.

It should be noted that being only to determine first software in multiple sample data in embodiments of the present invention Shared ratio for be illustrated, in another embodiment, if the negligible amounts of multiple sample datas, can also really Fixed first software ratio shared in all sample datas, the embodiment of the present invention is to without limitation.

2012: if the ratio is greater than or equal to preset ratio, it is determined that each in first software and multiple second softwares Promotion degree between second software, multiple second software are the software in multiple software in addition to first software, this is mentioned Whether liter degree is used to indicate first software has inhibition or castering action to the installation probability of other software.

Wherein, which can be arranged by the Server Default, can also by technical staff according to actual needs from Definition setting, it is not limited in the embodiment of the present invention.

If ratio determined by above-mentioned is greater than preset ratio, illustrate that the influence power of first software is larger, further needs Determine the quantity for the other software that first software is influenced.In order to determine the number of other software that first software is influenced Amount, it is thus necessary to determine which software first software has and influence, for this purpose, server determines first software and multiple second soft Promotion degree in part between each second software.

Wherein it is determined that the specific implementation of the promotion degree in first software and multiple second softwares between each second software It may include: the marginal probability of second software to be obtained for each second software, and determine and first software is being installed Under the conditions of the third condition probability of second software is installed, by the third condition probability divided by the marginal probability of second software, Obtain the promotion degree between first software and second software.

For example, the second software is L if first software is K, then the marginal probability P (L) of second software is obtained, and P (L | K) is counted, later, the promotion degree between the first software K and second software L can be obtained by P (L | K)/P (L).

It should be noted that it is above-mentioned be only with determine ratio be greater than or equal to corresponding first software of preset ratio with it is multiple It is illustrated for promotion degree in second software between each second software.In practical implementations, can also according to from greatly to Small sequence from multiple first softwares of n before selection percentage ranking in identified all first softwares, and determines the n the Promotion degree in one software in each first software and multiple second softwares between each second software, the embodiment of the present invention is to this Without limitation.

2013: software of the promotion degree less than 1 or greater than 1 in multiple second software of statistics between first software Quantity.

As it was noted above, if exist in multiple second software promotion degree between first software less than 1 or The second software greater than 1 then illustrates that first software plays inhibition or castering action to the installation probability of second software.Clothes Business device counts amount of software of the promotion degree less than 1 or greater than 1 in multiple second software between first software.

2014: when the amount of software of statistics reaches the preset quantity, determining that first software is designated software.

If the amount of software of statistics reaches the preset quantity, illustrate installation of first software to a large amount of second software There is probability promotion or inhibiting effect first software is therefore determined as designated software.According to this kind of implementation method Determine above-mentioned multiple designated softwares.

Step 202: determining the joint probability between the target software and identified designated software.

Due to will affect the target software when installation probability of the designated software to target software has promotion or inhibiting effect Software share accounting determination as a result, inhibiting or castering action accordingly, it is determined that having to the installation probability of target software After designated software, need to consider the Joint Distribution between the designated software and the target software, i.e. the server determines that this is specified Joint probability between software and the target software.

In the concrete realization, server obtains the marginal probability of identified designated software, and statistics is installing institute really The second condition probability of the target software is installed under conditions of fixed designated software, it is general based on the edge of identified designated software Rate and the second condition probability, determine the joint probability between the target software and identified designated software.

For example, being obtained if having the designated software of inhibition or castering action to the installation probability of target software Y is software X The marginal probability of the designated software is P (X), and obtaining installation under conditions of installing identified designated software by statistics should The second condition probability P (Y | X) of target software, then can determine combining between the target software and identified designated software Probability is P (X, Y)=P (X) * P (Y | X).

Since the marginal probability of the designated software is true probability, and also contemplate here specify determined by the installation it is soft The second condition probability of the target software is installed under conditions of part, therefore, inaccurate compared to the marginal probability of designated software Situation, the second condition probability of the marginal probability based on the designated software, the joint probability determined are corrected, that is, are corrected Influence of the designated software to the target software.

Step 203: under conditions of statistics designated software determined by installing and the target software at least two dimensions Corresponding terminal installs the first condition probability of the target software, and each dimension at least two dimension influences the target The software share accounting of software.

Software share accounting due to influencing the target software further includes at least two dimensions, in practical implementations, Also need to count the first condition probability that corresponding terminal at least two dimension installs the target software.Wherein, statistics The specific reality of the conditional probability of the target software is installed in specific implementation with above-mentioned statistics under conditions of being mounted with the designated software It is now similar.

For example, if at least two dimension includes terminal models and user's gender, P (type | X, Y) and P (property are counted Not | X, Y, type).In the concrete realization, since each dimension at least two dimension includes multiple dimensional informations, example Such as, terminal models include A and B, and user's gender includes male and female, therefore, in practical implementations, the P of statistics (type | X, Y) It also include multiple with P (gender | X, Y, type).For example, the first condition probability include P (A | X, Y), P (B | X, Y), P (male | X, Y, A), P (female | X, Y, A), P (male | X, Y, B) and P (female | X, Y, B).

Step 204: based between the target software and identified designated software joint probability and the first condition it is general Rate determines identified joint probability between designated software, the target software and at least two dimension.

In practical implementations, server based on the joint probability between the target software and identified designated software and is somebody's turn to do First condition probability, using Gibbs Sampling P (H, M)=P (H | M) * P (M), determine determined by designated software, the target Joint probability between software and at least two dimension.

Continue by taking above-mentioned example as an example, i.e., to the installation probability of target software X have promotion or inhibiting effect specify it is soft Part is software X, then server is using Gibbs Sampling P (H, M)=P (H | M) * P (M), can determine designated software X, should Joint probability between target software Y and at least two dimension includes:

Step 205: general based on the joint between identified designated software, the target software and at least two dimension The dimensional information of rate, multiple sample datas of storage and multiple sample data obtains the software share accounting of the target software.

In the concrete realization, dimensional information of the server based on the joint probability and multiple sample datas, from storage The sample data of specified quantity is extracted in multiple sample datas, wherein specified quantity can be made by oneself according to actual needs by user Justice setting, can also be arranged, it is not limited in the embodiment of the present invention by the Server Default.

For example, it is assumed that the specified quantity is m, then the server is more from this according to the dimensional information of multiple sample data Sample size extraction is carried out in a sample data, * m dimensional information is A, male, and is equipped with for example, extracting P (X, Y, A, male) The sample data that the target software Y terminal for having influential designated software X is reported；Extract P (X, Y, B, male) * m dimension letter Breath is B, male, and is equipped with the sample data etc. reported on the target software Y terminal for having influential designated software X.

Later, server is the software share accounting that can determine the target software in m extracted sample data.Example Such as, t sample data is corresponding with the target software in m sample data, then the software share accounting is (t/m) * 100%.

Fig. 2 B is please referred to, which is a kind of correction result schematic diagram shown according to an exemplary embodiment.Wherein, institute It is X, target software Y with determining designated software, and it is soft to belong to brother between identified designated software X and target software Y Part product, i.e., identified designated software X have castering action to the installation probability of target software Y.It can from Fig. 2 B To find out, higher problem caused by software share accounting of the X to Y is had modified by the above method.

Fig. 2 C is please referred to, Fig. 2 C is a kind of correction result schematic diagram shown according to another exemplary embodiment, wherein institute Determining designated software is X, target software Y, and it is soft to belong to competition between identified designated software X and target software Y Part product, i.e., identified designated software X are inhibited to the installation probability of target software Y.It can from Fig. 2 C To find out, relatively low problem caused by software share accounting of the X to Y is had modified by the above method.

Based on the joint probability between identified designated software, the target software and at least two dimension, storage Multiple sample datas and multiple sample data dimensional information, before the software share accounting for counting the target software, need Determine the dimensional information of multiple sample datas and multiple sample datas, specific implementation includes: to receive multiple data sources to report Sample data and the sample data dimensional information, according to the weight of multiple data source, from received all sample datas The dimensional information of middle determination multiple sample data and multiple sample data.

In the concrete realization, the dimension that server receives sample data and the sample data that multiple data sources report is believed Breath determines multiple sample data and multiple sample according to the weight of multiple data source from received all sample datas The dimensional information of notebook data.

For example, can extract more part from received all sample datas should if the weight of certain data source is larger The sample data that data source reports.Conversely, can be taken out from received all sample datas if the weight of certain data source is smaller The sample data for taking the fewer parts data source to report.In this way, multiple sample data can be obtained.

Wherein, the dimensional information of each sample data can be by data source from such as QQ, micro- in above-mentioned multiple sample datas It acquires, and is reported in the server in the social category software of letter etc.

It should be noted that the weight of multiple data source can be configured according to the actual situation by technical staff, example Such as, if data source is by sofeware management reported data sample, and portion is larger in the market for the sofeware management, then can incite somebody to action The weight of the data source is arranged larger.The specific setting rule of weight about each data source, the embodiment of the present invention It does not limit this.

In embodiments of the present invention, it determines that there is the installation probability of target software from multiple designated softwares and inhibits or mention The designated software of effect is risen, and determines the joint probability between the target software and the designated software.Later, statistics is in installation institute Under conditions of determining designated software and target software first of corresponding terminal installation targets software at least two dimensions Part probability determines institute based on the joint probability and the first condition probability between the target software and identified designated software Joint probability between determining designated software, the target software and at least two dimension, based on the joint probability and more The dimensional information of a sample data and multiple sample datas determines software share accounting.In the process, it is contemplated that designated software And at least two conditional relationship between dimension and influence the Joint Distribution of the target software, ensure that determining software share The accuracy of accounting, also, due to being not necessarily to industry background veteran data system during determining software share accounting Meter personnel are manually participated in, and this improves determine efficiency.

Next, including here such as above-mentioned terminal models, user's gender, age of user with influence software share accounting Etc at least two dimensions for be illustrated, refer to Fig. 3 A, which is a kind of software provided in an embodiment of the present invention The flow chart of the determination method of share accounting, this method are applied in server, and the determination method of the software share accounting includes:

Step 301: obtain target dimension on corresponding terminal installation targets software marginal probability, the target software be to The software of statistical software share accounting, the target dimension are to influence at least two dimensions of the software share accounting of the target software In with proper boundary probability dimension.

In practical implementations, which can be preset by technical staff, that is to say, technical staff can basis The authenticity of the marginal probability of actual storage determines the target dimension from least two dimension.

In one possible implementation, Fig. 3 B is please referred to, can store in the server corresponding in each dimension Terminal installation targets software marginal probability.In that case, which can be from each dimension that itself is stored In the marginal probability of corresponding terminal installation targets software, the side of corresponding terminal installation targets software in the target dimension is obtained Edge probability.

Since for each dimension at least two dimension, which is corresponding with multiple dimensional informations, for example, with The dimension is for user's gender, the corresponding dimensional information of the dimension includes male and female is using the dimension as terminal models for another example Example, the corresponding dimensional information of the dimension may include millet, Huawei, three magnitudes.Therefore, in practical implementations, for the target For dimension, the marginal probability for needing to obtain includes the marginal probability of the corresponding multiple dimensional informations of the target dimension.

For example, it is assumed that the target dimension is terminal models, and the corresponding dimensional information of the terminal models includes type A and machine Type B, then the server obtains the corresponding terminal of type A and installs the marginal probability P (A) of the target software, and obtains B pairs of type The terminal answered installs the marginal probability P (B) of the target software.

It should be noted that above-mentioned is only soft to be stored with corresponding terminal installation targets in each dimension in the server It is illustrated for the marginal probability of part.In another embodiment, which can also provide from other third party partners Server in obtain the marginal probability of corresponding terminal installation targets software in the target dimension, the embodiment of the present invention to this not It limits.

Step 302: this is at least under conditions of terminal corresponding in the target dimension has installed the target software for statistics Corresponding terminal installs the fourth condition probability of the target software in other dimensions in two dimensions.

In practical implementations, due to be not between each dimension in above-mentioned at least two dimension it is independent of each other, this The relationship between each dimension is considered in inventive embodiments, i.e. statistics corresponding terminal in the target dimension has installed this Corresponding terminal installs the fourth condition of the target software in other dimensions at least two dimension under conditions of target software Probability.

In the concrete realization, when only existing two dimensions influences software share accounting, e.g., which includes terminal Type and user's gender, the server statistics corresponding terminal in this dimension of terminal models have installed the target software Under the conditions of, corresponding terminal installs the fourth condition probability of the target software in this dimension of user's gender.

When there are more than two dimensions influences software share accounting, for example, it is assumed that the two or more dimension includes terminal Type, user's gender and age of user, the server statistics corresponding terminal in this dimension of terminal models have installed this Under conditions of target software, corresponding terminal installs the fourth condition of the target software in user's gender and age of user dimension Probability, also, in the concrete realization, the fourth condition probability include, P (user's gender | terminal models) and P (age of user | it uses Family gender, terminal models).

In order to make it easy to understand, being illustrated followed by specific example.Assuming that the corresponding dimension of this dimension of age of user Spending information includes greater than 30 years old and being less than or equal to 30 years old, then in practical implementations, server needs the fourth condition counted Probability includes P (male | A), P (female | A), P (> 30 | male, A), P (≤30 | male, A), P (> 30 | female, A), P (≤30 | female, A), P (male | B), P (female | B), P (> 30 | male, B), (≤30 P | male, B), P (> 30 | female, B) and (≤30 P | female, B).

It should be noted that in the concrete realization, which can be based on all sample numbers that each data source reports According to each conditional probability for including to above-mentioned fourth condition probability counts.

It should also be noted that, being only to determine that terminal corresponding in the target dimension has been pacified in the method for statistics here Fill under conditions of the target software that corresponding terminal installs the 4th of the target software in other dimensions at least two dimension It is illustrated for conditional probability, in another embodiment, can also be directly obtained by other channels in the target dimension Upper corresponding terminal has installed under conditions of the target software at least two dimension corresponding terminal peace in other dimensions The fourth condition probability of the target software is filled, for example, can be acquired from some files that official issues by technical staff After be stored in the server, it is not limited in the embodiment of the present invention.

Step 303: based on the marginal probability and the fourth condition probability, determining that the joint between at least two dimension is general Rate.

In the concrete realization, server can be sampled former based on the marginal probability and the fourth condition probability using Gibbs Reason P (H, M)=P (H | M) * P (M) determines the joint probability between at least two dimension.

Continue by taking above-mentioned example as an example, i.e., corresponding terminal installs the marginal probability of the target software in the target dimension Including P (A) and P (B), and the fourth condition conditional probability includes P (male | A), P (female | A), P (> 30 | male, A), (≤30 P | male, A), P (> 30 | female, A), (≤30 P | female, A), P (male | B), P (female | B), P (> 30 | male, B), (≤30 P | male, B), P (> 30 | Female, B) and (≤30 P | female, B), then server uses above-mentioned Gibbs Sampling, can determine between multiple dimension Joint probability includes:

Since the marginal probability that the terminal in target dimension installs the target software is accurate marginal probability, and in determination When joint probability between above-mentioned at least two dimension, it is contemplated that the conditional relationship between each dimension, accordingly, it is determined that acquired Joint probability it is more accurate, so, it is ensured that the accuracy of the software share accounting of subsequent determination.

Step 304: the dimensional information based on the joint probability, multiple sample datas of storage and multiple sample data, Obtain the software share accounting of the target software.

For example, it is assumed that the specified quantity is m, then the server is more from this according to the dimensional information of multiple sample data Extract respectively in a sample data: P (A, male, > 30) * m dimensional information is A, the sample data that male and > 30 terminals report；P (A, female, > 30) * m dimensional information is A, female and > 30 the sample data that reports of terminal；P (A, male ,≤30) * m dimension letter Breath is A, the sample data that the terminal of Ji≤30 Nan reports；P (,≤30 A, Nv) * m dimensional information is A, the terminal of Ji≤30 Nv The sample data reported；P (B, male, > 30) * m dimensional information is B, the sample data that male and > 30 terminals report；P (B, female, ﹥ 30) * m dimensional information is B, female and > 30 the sample data that reports of terminal；P (,≤30 B, Nan) * m dimensional information is B, The sample data that the terminal of Ji≤30 Nan reports；And P (,≤30 B, Nv) * m dimensional information is B, in the terminal of Ji≤30 Nv The sample data of report.

Later, in the m sample data that server statistics are extracted, the ratio of share shared by the target software is to get arriving The software share accounting of the target software.

Fig. 3 C is please referred to, which shows the effect diagram of the software share accounting determined based on terminal models, from In Fig. 3 C as can be seen that for the situation (as shown in 11 in Fig. 3 C) for causing software share accounting higher as terminal models, Reduce software share accounting (as shown in 12 in Fig. 3 C), i.e., software share accounting obtains Amendment.

Further, please refer to Fig. 3 B, based on the joint probability, storage multiple sample datas and multiple sample number According to dimensional information, before the software share accounting for counting the target software, server is also it needs to be determined that multiple sample data With the dimensional information of multiple sample datas.

Specifically, the dimensional information for receiving sample data and the sample data that multiple data sources report, according to multiple The weight of data source determines the dimension letter of multiple sample data and multiple sample data from received all sample datas Breath.

In addition, please referring to Fig. 3 B, in practical implementations, after determining software share accounting, can be serviced by big data flat Platform carries out alternate analysis verifying to it, and it is not limited in the embodiment of the present invention for specific verification process.Also, it is soft after verifying Part share, which accounts for, can be obtained actual application, for example, can be applied to algorithm fix, historical data tracing, specific analysis test Etc. in scenes.

In embodiments of the present invention, the target that corresponding terminal in target dimension installs software share accounting to be counted is obtained The marginal probability of software, and count under conditions of terminal corresponding in target dimension has installed the target software at least Corresponding terminal installs the fourth condition probability of the target software in other dimensions in two dimensions.Later, general based on the edge Rate and fourth condition probability determine the joint probability between at least two dimension, since the target dimension is to influence the target In at least two dimensions of the software share accounting of software with proper boundary probability dimension, and consider the target dimension with Relationship between other dimensions, therefore, obtained joint probability are more accurate.In this way, when based on the joint probability, storage it is more The dimensional information of a sample data and multiple sample data, when counting the software share accounting of the target software, it is ensured that The accuracy of the software share accounting of statistics.Also, due to carrying out people without the veteran data statistics personnel of industry background Work point analysis and amendment, this improves the determination efficiency of software share accounting.

A referring to fig. 4, Fig. 4 A are that the structure of the determining device of another software share accounting provided in an embodiment of the present invention is shown It is intended to, which includes: the first determining module 401, the second determining module 402, the first statistical module 403, third determining module 404 and the 4th determining module 405；

First determining module 401, for executing the step 201 in above-mentioned Fig. 2A embodiment；

Second determining module 402, for executing the step 202 in above-mentioned Fig. 2A embodiment；

First statistical module 403, for executing the step 203 in above-mentioned Fig. 2A embodiment；

Third determining module 404, for executing the step 204 in above-mentioned Fig. 2A embodiment；

4th determining module 405, for executing the step 205 in above-mentioned Fig. 2A embodiment.

Optionally, Fig. 4 B is please referred to, the device further include:

5th determining module 406, for executing the step 2011 in above-mentioned Fig. 2A embodiment；

6th determining module 407, for executing the step 2012 in above-mentioned Fig. 2A embodiment；

Second statistical module 408, for executing the step 2013 in above-mentioned Fig. 2A embodiment；

7th determining module 409, for executing the step 2014 in above-mentioned Fig. 2A embodiment.

Referring to Fig. 5 A, a kind of Fig. 5 A structural representation of the determining device of software share accounting provided in an embodiment of the present invention Figure, the device include obtaining module 501, statistical module 502, the first determining module 503 and the second determining module 504.

Module 501 is obtained, for executing the step 301 in above-mentioned Fig. 3 A embodiment；

Statistical module 502, for executing the step 302 in above-mentioned Fig. 3 A embodiment；

First determining module 503, for executing the step 303 in above-mentioned Fig. 3 A embodiment；

Second determining module 504, for executing the step 304 in above-mentioned Fig. 3 A embodiment.

Optionally, referring to Fig. 5 B, the device further include:

Receiving module 505, for receiving the dimensional information of sample data and the sample data that multiple data sources report；

Third determining module 506, for the weight according to the multiple data source, from received all sample datas really The dimensional information of fixed the multiple sample data and the multiple sample data.

It should be understood that the determining device of software share accounting provided by the above embodiment is realizing software share accounting Determination method when, only the example of the division of the above functional modules, in practical application, can according to need and incite somebody to action Above-mentioned function distribution is completed by different functional modules, i.e., the internal structure of device is divided into different functional modules, with complete At all or part of function described above.In addition, the determining device of software share accounting provided by the above embodiment with it is soft The determination embodiment of the method for part share accounting belongs to same design, and specific implementation process is detailed in embodiment of the method, here no longer It repeats.

Fig. 6 is a kind of server architecture schematic diagram of the determining device of software share accounting provided in an embodiment of the present invention. The server can be the server in background server cluster.Specifically:

Server 600 includes central processing unit (CPU) 601 including random access memory (RAM) 602 and read-only deposits The system storage 604 of reservoir (ROM) 603, and the system bus of connection system storage 604 and central processing unit 601 605.Server 600 further includes the basic input/output (I/O that information is transmitted between each device helped in computer System) 606, and for the mass-memory unit of storage program area 613, application program 614 and other program modules 615 607。

Basic input/output 606 includes display 608 for showing information and inputs information for user The input equipment 609 of such as mouse, keyboard etc.Wherein display 608 and input equipment 609 are all by being connected to system bus 605 input and output controller 610 is connected to central processing unit 601.Basic input/output 606 can also include defeated Enter o controller 610 for receiving and handling from the defeated of multiple other equipment such as keyboard, mouse or electronic touch pen Enter.Similarly, input and output controller 610 also provides output to display screen, printer or other kinds of output equipment.

Mass-memory unit 607 is connected by being connected to the bulk memory controller (not shown) of system bus 605 To central processing unit 601.Mass-memory unit 607 and its associated computer-readable medium are that server 600 provides Non-volatile memories.That is, mass-memory unit 607 may include such as hard disk or CD-ROM drive etc Computer-readable medium (not shown).

Without loss of generality, computer-readable medium may include computer storage media and communication media.Computer storage Medium includes any of the information such as computer readable instructions, data structure, program module or other data for storage The volatile and non-volatile of method or technique realization, removable and irremovable medium.Computer storage medium include RAM, ROM, EPROM, EEPROM, flash memory or other solid-state storages its technologies, CD-ROM, DVD or other optical storages, cassette, magnetic Band, disk storage or other magnetic storage devices.Certainly, skilled person will appreciate that computer storage medium is not limited to It states several.Above-mentioned system storage 604 and mass-memory unit 607 may be collectively referred to as memory.

According to various embodiments of the present invention, server 600 can also pass through the network connections such as internet to network On remote computer operation.Namely server 600 can be by the Network Interface Unit 611 that is connected on system bus 605 It is connected to network 612, in other words, Network Interface Unit 611 can be used also to be connected to other kinds of network or long-range meter Calculation machine system (not shown).

Above-mentioned memory further includes one, and perhaps more than one program one or more than one program are stored in storage In device, it is configured to be executed by CPU.The one or more programs include provided in an embodiment of the present invention soft for carrying out The instruction of the determination method of part share accounting.

In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instruction, example are additionally provided It such as include the memory 604 of instruction, above-metioned instruction can be executed by the processor 620 of device 600 to complete the above method.For example, The non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk With optical data storage devices etc..

A kind of non-transitorycomputer readable storage medium, when the instruction in the storage medium is by the processor of server When execution, the determination method for executing software share accounting described in above-mentioned Fig. 2A or Fig. 3 A is enabled the server to.

Those of ordinary skill in the art will appreciate that realizing that all or part of the steps of above-described embodiment can pass through hardware It completes, relevant hardware can also be instructed to complete by program, the program can store in a kind of computer-readable In storage medium, storage medium mentioned above can be read-only memory, disk or CD etc..

The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.

Claims

1. a kind of determination method of software share accounting, which is characterized in that the described method includes:

Determine that the installation probability to target software has the designated software of inhibition or castering action from multiple designated softwares, each Designated software refers to that the installation probability to preset quantity other software all has the software of inhibition or castering action, the target Software is the software of software share accounting to be counted；

Corresponding terminal at least two dimensions under conditions of statistics designated software determined by installing and the target software The first condition probability of the target software is installed, each dimension at least two dimension influences the target software Software share accounting；

Based between the target software and identified designated software joint probability and the first condition probability, determine institute Joint probability between determining designated software, the target software and at least two dimension；

Based between identified designated software, the target software and at least two dimension joint probability, storage The dimensional information of multiple sample datas and the multiple sample data obtains the software share accounting of the target software.

2. the method as described in claim 1, which is characterized in that the determination target software and identified designated software Between joint probability, comprising:

The marginal probability of designated software determined by obtaining, and installed under conditions of statistics designated software determined by installing The second condition probability of the target software；

Based on the marginal probability and the second condition probability of identified designated software, determines the target software and determine Designated software between joint probability.

3. method according to claim 1 or 2, which is characterized in that described to determine from multiple designated softwares to the target The installation probability of software has before the designated software of inhibition or castering action, further includes:

Determine the first software ratio shared in the multiple sample data, first software is the multiple sample data Any software in corresponding multiple softwares；

If the ratio is greater than or equal to preset ratio, it is determined that first software and in multiple second softwares each second soft Promotion degree between part, the multiple second software is the software in the multiple software in addition to first software, described Whether promotion degree is used to indicate first software has inhibition or castering action to the installation probability of other software；

Count amount of software of the promotion degree in the multiple second software between first software less than 1 or greater than 1；

When the amount of software of statistics reaches the preset quantity, determine that first software is designated software.

4. method as claimed in claim 3, which is characterized in that every in determination first software and multiple second softwares Promotion degree between a second software, comprising:

For each second software, the marginal probability of second software is obtained, and determines and first software is being installed Under the conditions of the third condition probability of second software is installed；

By the third condition probability divided by the marginal probability of second software, first software and described second soft is obtained Promotion degree between part.

5. the method as described in Claims 1-4 is any, which is characterized in that described based on identified designated software, the mesh Mark joint probability, the dimension of multiple sample datas of storage and the multiple sample data between software and at least two dimension Information is spent, before obtaining the software share accounting of the target software, further includes:

Receive the dimensional information of sample data and the sample data that multiple data sources report；

According to the weight of the multiple data source, the multiple sample data and described is determined from received all sample datas The dimensional information of multiple sample datas.

6. a kind of determination method of software share accounting, which is characterized in that the described method includes:

The marginal probability of corresponding terminal installation targets software in target dimension is obtained, the target software is software part to be counted The software of volume accounting, the target dimension be the software share accounting for influencing the target software at least two dimensions in have The dimension of proper boundary probability；

Statistics at least two dimension under conditions of terminal corresponding in the target dimension has installed the target software Corresponding terminal installs the fourth condition probability of the target software in other dimensions in degree；

Based on the marginal probability and the fourth condition probability, the joint probability between at least two dimension is determined；

Dimensional information based on the joint probability, multiple sample datas of storage and the multiple sample data obtains described The software share accounting of target software.

7. method as claimed in claim 6, which is characterized in that multiple sample numbers based on the joint probability, storage According to the dimensional information with the multiple sample data, before obtaining the software share accounting of the target software, further includes:

8. a kind of determining device of software share accounting, which is characterized in that described device includes:

First determining module inhibits or is promoted to make for determining to have the installation probability of target software from multiple designated softwares Designated software, each designated software, which refers to that the installation probability of preset quantity other software is all had inhibition or promoted, to be made Software, the target software are the software of software share accounting to be counted；

First statistical module, at least two under conditions of for counting the designated software determined by installing and the target software Corresponding terminal installs the first condition probability of the target software in dimension, and each dimension at least two dimension is equal Influence the software share accounting of the target software；

Third determining module, for based on the joint probability and described between the target software and identified designated software One conditional probability determines identified joint probability between designated software, the target software and at least two dimension；

4th determining module, for based between identified designated software, the target software and at least two dimension Joint probability, the dimensional information of multiple sample datas of storage and the multiple sample data, obtain the target software Software share accounting.

9. device as claimed in claim 8, which is characterized in that second determining module is used for:

10. device as claimed in claim 8 or 9, which is characterized in that described device further include:

5th determining module, for determining the first software ratio shared in the multiple sample data, first software For any software in the corresponding multiple softwares of the multiple sample data；

6th determining module, if being greater than or equal to preset ratio for the ratio, it is determined that first software and multiple the Promotion degree in two softwares between each second software, the multiple second software are in the multiple software except described first is soft Software except part, the promotion degree be used to indicate first software to the installation probability of other software whether have inhibition or Castering action；

Second statistical module, for count the promotion degree in the multiple second software between first software less than 1 or Person is greater than 1 amount of software；

7th determining module refers to for when the amount of software of statistics reaches the preset quantity, determining first software Determine software.

11. a kind of determining device of software share accounting, which is characterized in that described device includes:

Module is obtained, for obtaining the marginal probability of corresponding terminal installation targets software in target dimension, the target software For the software of software share accounting to be counted, the target dimension is to influence the software share accounting of the target software at least With the dimension of proper boundary probability in two dimensions；

Statistical module, for counting corresponding terminal under conditions of of having installed target software institute in the target dimension State at least two dimensions that corresponding terminal installs the third condition probability of the target software in other dimensions；

First determining module determines at least two dimension for being based on the marginal probability and the third condition probability Between joint probability；

Second determining module, for multiple sample datas and the multiple sample data based on the joint probability, storage Dimensional information obtains the software share accounting of the target software.

12. device as claimed in claim 11, which is characterized in that described device further include:

Receiving module, for receiving the dimensional information of sample data and the sample data that multiple data sources report；

Third determining module, for the weight according to the multiple data source, from received all sample datas described in determination The dimensional information of multiple sample datas and the multiple sample data.

13. a kind of terminal, which is characterized in that the terminal includes processor and memory, is stored at least in the memory One instruction, at least one section of program, code set or instruction set, described instruction, described program, the code set or described instruction collection Loaded by the processor and executed the determination method to realize the software share accounting as described in claim 1 to 7 is any.

14. a kind of computer readable storage medium, which is characterized in that be stored at least one instruction, extremely in the storage medium Few one section of program, code set or instruction set, described instruction, described program, the code set or described instruction collection are added by processor Carry and execute the determination method to realize the software share accounting as described in claim 1 to 7 is any.