KR101871739B1 - Method of determining representative keyword of application and system for determining representative keyword of application - Google Patents

Abstract

The present invention relates to a method for determining a representative keyword of an application, which comprises the steps of: crawling description data of a plurality of applications from an application market; determining keywords included in the description data and the importance of the keywords in the description data to store the same in a first database, with respect to each of the applications; selecting a target application among the applications; obtaining a list of associated applications having a high correlation with the target application among the applications; reading each keyword of the target application and the associated applications and the importance of the keywords from the first database; determining the keywords of the target application overlapping with the keywords of the associated applications, as redundant keywords; and determining representative keywords representing the target application among the keywords of the target application on the basis of the importance of the keywords of the target application and the importance of the redundant keywords of each of the associated applications.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a representative keyword determination method for an application, and a representative keyword determination system for an application. [0002]

BACKGROUND OF THE INVENTION 1. Field of the Invention [0002] The present invention relates to a technique for analyzing keywords representing mobile applications, and more particularly, to a representative keyword determination method for an application and a representative keyword determination system for an application.

Recently, as the performance of mobile terminals such as smart phones and tablet PCs has improved and wireless communication speeds have increased, the kinds of applications that can be executed in mobile terminals have been rapidly increasing.

Therefore, it is increasingly difficult for mobile terminal users to find an application desired by a user among a large number of applications.

Therefore, Open Market, which sells application development companies and applications, provides explanatory materials showing the characteristics, characteristics, and functions of applications with applications.

In an open market for selling applications, recently, applications are classified into a plurality of categories according to their types, and category information of an application is provided to a user together with explanatory data of the application, thereby making it easy to improve user's convenience of searching for an application .

However, it is difficult for the user to understand the characteristics and functions of the application only by the category information of the application. Also, by reading explanatory data of various applications to find the application desired by the user, There is a problem that effort is required.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method and system for determining representative keywords of an application that can effectively represent a feature of a mobile application.

In order to accomplish one object of the present invention, in a representative keyword determination method of an application according to an embodiment of the present invention, crawling description data of each of a plurality of applications from an application market, For each of the plurality of applications, the weight of the keywords in the keywords included in the explanatory data and the explanatory data is determined and stored in the first database, the target application is selected from among the plurality of applications, Obtaining a list of associated applications having a high degree of association with the target application, reading the weight of the keywords and the keywords of each of the target application and the associated applications from the first database, Determining, based on the weight of the keywords of the target application and the weight of the duplicate keywords of each of the associated applications, determining keywords that overlap with the keywords of the related applications among the keywords of the target applications, And determines representative keywords representing the target application from among the keywords of the target application.

In one embodiment, for each of the plurality of applications, the step of determining the weight of the keywords in the keywords and the explanatory data included in the explanatory data and storing the determined weight in the first database may include: Performing morphological analysis on the explanatory data for each of the applications to extract the keywords included in the explanatory data; determining a repetition frequency of each of the keywords in the explanatory data; Determining, by the weight of each of the keywords, a value obtained by dividing the repetition count of each of the keywords by the sum of the repetition counts of the keywords, and associating the weight of the keywords and the keywords with the name of the corresponding application To the first database It can be included.

In one embodiment, the step of obtaining a list of the associated applications having a high degree of association with the target application among the plurality of applications may include, for each of the plurality of applications, And reading a list of the associated applications corresponding to the target application from a second database that stores a list of the associated applications in advance.

In one embodiment, the step of obtaining a list of the associated applications having a high degree of association with the target application among the plurality of applications comprises the steps of: obtaining an identification ID of each of the plurality of mobile terminals, Receiving install log data including a list of installed applications and storing the received install log data in a third database, storing installation log data including a list of installed applications in a first database, Determining k-th total ratios corresponding to the ratios of mobile terminals in which each k-th (k is a positive integer) application is installed; Based on the target application Determining k-th intersection ratios corresponding to ratios of mobile terminals in which each of the first to k-th applications is installed among mobile terminals with installed first to k-th applications, Determining k-th association indices indicative of a degree of association between the target application and each of the first to k-th applications based on the k-th to k-th crossing ratios, Determining the applications with a relatively high association index as the associated applications.

The first to kth applications corresponding to the ratios of mobile terminals installed with the first to k-th applications among the plurality of mobile terminals, based on the installation log data of the plurality of mobile terminals stored in the third database, Determining all ratios includes determining a total number of the identification IDs corresponding to the plurality of mobile terminals based on the installation log data of the plurality of mobile terminals, Determining, as a j-th installation number, the number of identification IDs corresponding to mobile terminals having a j-th application (where j is any positive integer equal to or greater than 1 k) based on installation log data; and Determining the value obtained by dividing the installation number by the total number as the total j ratio System.

The first to the kth applications corresponding to the ratio of the mobile terminals installed with the first to k-th applications among the mobile terminals installed with the target application on the basis of the installation log data of the plurality of mobile terminals stored in the third database, Determining the kth intersection ratios comprises: determining a number of the identification IDs corresponding to mobile terminals on which the target application is installed based on the installation log data of the plurality of mobile terminals as a target installation number, Based on the installation log data of the mobile terminals of the mobile terminals of the mobile terminals, the number of the identification IDs corresponding to the mobile terminals in which the target application and the j (j is an arbitrary positive integer equal to or greater than 1) Determining a number, Wherein the j-th cross-number can include determining a value obtained by dividing the number of target installed in the j-th cross ratio.

The first through k-th associative exponents representing the degree of association between the target application and each of the first through k-th applications based on the first through k-th overall ratios and the first through k- Determining may comprise determining the first to k < RTI ID = 0.0 > k < / RTI > associated exponents based on a difference between each of the first through k total ratios and each of the first through k crossings ratios.

The first through k-th associative exponents representing the degree of association between the target application and each of the first through k-th applications based on the first through k-th overall ratios and the first through k- The determining may comprise determining the first through k k -th associative exponents based on a ratio of each of the first through k-th crossing ratios for each of the first through k-th overall ratios.

The first through k-th associative exponents representing the degree of association between the target application and each of the first through k-th applications based on the first through k-th overall ratios and the first through k- Determining the difference between each of said first through kth total ratios and each of said first through kth intersection ratios and said first through kth intersection ratios for each of said first through k total ratios, K < / RTI > associated indexes based on the ratio of each of the first through k < th >

The first through k-th associative exponents representing the degree of association between the target application and each of the first through k-th applications based on the first through k-th overall ratios and the first through k- Determining a difference between each of said first through kth total ratios and each of said first through kth intersection ratios, said first through kth intersection ratios for each of said first through kth total ratios, , And determining the first through k < RTI ID = 0.0 > k < / RTI > associated exponents based on a weight having a ratio of each of the first through k and a value greater than zero and less than one.

In one embodiment, the representative keywords representing the target application among the keywords of the target application are determined based on the weight of the keywords of the target application and the weight of the duplicate keywords of each of the associated applications For each of the duplicate keywords among the keywords of the target application, applying a weight of the duplicate keyword of each of the associated applications to a weight of the duplicate keyword of the target application as a weight, Determining a score of the non-overlapping keyword in the target application for each of the non-overlapping keywords that are not duplicate keywords among the keywords in the target application System, and it can include determining a keyword having a relatively high score among the keyword of the target application with the representative keywords.

For each of the duplicate keywords among the keywords of the target application, a weight of the duplicate keyword is determined by applying a weight of the duplicate keyword of each of the associated applications to a weight of the duplicate keyword of the target application May include determining a sum of the weights of the duplicate keywords of the associated applications and a specific gravity of the duplicate keywords of the target application as a score of the duplicate keyword.

In one embodiment, the step of acquiring a list of the associated applications having a high degree of association with the target application among the plurality of applications may include acquiring, from among the plurality of applications, Obtaining a list of the associated applications corresponding to the target application and associated indexes representing the degree of association between the target application and the associated applications.

Wherein the step of determining the representative keywords representing the target application among the keywords of the target application based on the weight of the keywords of the target application and the weight of the overlapping keywords of each of the associated applications, For each of the duplicate keywords among the keywords of the application, the weight of the duplicate keyword of each of the associated applications and the association index of each of the associated applications is applied as a weight to the weight of the duplicate keyword of the target application Determining a score of the duplicate keyword, calculating, for each of the non-duplicated keywords that are not duplicate keywords among the keywords of the target application, Determining a score of the keyword clothing, and may include determining keywords having a relatively high score among the keyword of the target application with the representative keywords.

For each of the duplicate keywords among the keywords of the target application, a weight of the duplicate keyword of each of the associated applications and a weight of the association index of each of the associated applications in the weight of the duplicate keyword of the target application Wherein the step of determining the score of the duplicate keyword comprises: calculating a sum of the weight of the duplicate keyword of each of the associated applications and the association index of each of the associated applications together with the specific gravity of the duplicate keyword of the target application, As a score of a duplicate keyword.

In one embodiment, the representative keywords representing the target application among the keywords of the target application are determined based on the weight of the keywords of the target application and the weight of the duplicate keywords of each of the associated applications Determining a score for each of the keywords in the target application based on the weight of the keywords in the target application and the weight of the duplicate keywords in each of the associated applications, Determining a correction coefficient for each of the keywords of the target application based on the number of applications including keywords overlapping each of the keywords of the target application, Of the target application by the correction coefficient of each of the keywords of the target application as the correction score of each of the keywords of the target application, And determining the keywords having the correction score as the representative keywords.

The step of determining the correction coefficient of each of the keywords of the target application based on the number of applications including a keyword overlapping each of the keywords of the target application among the plurality of applications comprises: Determining the correction coefficient of the keyword of the target application to be a relatively small value when the number of applications including the same keyword as the keyword of the target application is relatively large among the plurality of applications, Determining the correction coefficient of the keyword of the target application as a relatively large value when the number of applications including the same keyword as the keyword of the target application is relatively small, .

In one embodiment, the representative keyword determination method of the application may further include storing category information of each of the plurality of applications in the first database based on the description data of each of the plurality of applications have. Wherein the step of determining the representative keywords representing the target application among the keywords of the target application based on the weight of the keywords of the target application and the weight of the overlapping keywords of each of the associated applications, Determining a score of each of the keywords in the target application based on the weight of the keywords in the application and the weight of the duplicate keywords in each of the associated applications; Determining a correction coefficient for each of the keywords of the target application based on the number of categories to which the applications that contain the keywords overlap each other, Of the target application by the correction coefficient of each of the keywords of the target application as the correction score of each of the keywords of the target application, And determining the keywords having the correction score as the representative keywords.

The step of determining the correction coefficient of each of the keywords of the target application based on the number of categories to which the applications including the keywords overlapping each of the keywords of the target application among the plurality of applications belongs, Determining the correction coefficient of the keyword of the target application to be a relatively small value when the number of categories to which the applications including the same keyword as the keyword of the target application belongs is relatively large among the applications of the target application, When the number of categories to which the applications including the same keyword as the keyword of the target application among the plurality of applications belongs is relatively small, The correction coefficient of the word may include the step of relatively the greater the value.

In order to accomplish one object of the present invention, a representative keyword determination system of an application according to an embodiment of the present invention includes a database, a description data collection unit, a keyword extraction unit, an associated application determination unit, and a keyword analysis unit. The description data collection unit crawls description data of each of a plurality of applications from an application market. The keyword extracting unit determines the specific weight of the keywords in the description data and the description data for each of the plurality of applications and stores the determined weight in the database. The associated application determination unit selects a target application among the plurality of applications, and acquires a list of related applications having a high degree of association with the target application among the plurality of applications. Wherein the keyword analyzing unit reads the weightings of the keywords and the keywords of each of the target application and the related applications from the database and stores keywords overlapping the keywords of the related applications among the keywords of the target application Determines representative keywords representing the target application among the keywords of the target application based on the weight of the keywords of the target application and the weight of the duplicate keywords of each of the associated applications .

The representative keyword determination system of the application and the representative keyword determination method of the application according to the embodiments of the present invention can accurately determine the representative keywords that effectively characterize the application.

1 is a diagram illustrating a representative keyword determination system of an application according to an exemplary embodiment of the present invention.
2 is a flowchart illustrating a representative keyword determination method of an application according to an embodiment of the present invention.
FIG. 3 is a flowchart illustrating an example of determining the weight of the keywords in the description data and the keywords in the description data for each of the plurality of applications in FIG. 2 and storing the determined weight in the first database.
4 is a diagram showing an example of a first database included in the representative keyword determination system of the application of FIG.
5 is a diagram showing an example of a representative keyword determination system of the application shown in FIG.
6 is a diagram showing an example of a second database included in the representative keyword determination system of the application of FIG.
7 is a block diagram showing an example of an associated application determination unit included in the representative keyword determination system of the application shown in FIG.
FIG. 8 is a flowchart showing an example of a step of acquiring a list of associated applications having a high degree of association with a target application among the plurality of applications of FIG. 2. FIG.
9 is a diagram illustrating an example of installation log data transmitted from each of a plurality of mobile terminals shown in FIG. 7 to an associated application determination unit.
10 is a diagram showing an example of a third database included in the associated application determination unit of FIG.
FIG. 11 is a flowchart showing an example of determining the first to k-th overall ratios corresponding to the ratios of mobile terminals in which each of the first to k-th applications among the plurality of mobile terminals in FIG. 8 is installed.
FIG. 12 is a flowchart showing an example of determining the first through k-th intersection ratios corresponding to the ratios of the mobile terminals installed with the first through k-th applications among the mobile terminals installed with the target application of FIG.
13 shows an example of a step of determining representative keywords representing the target application among the keywords of the target application on the basis of the weight of the keywords of the target application and the weight of the duplicate keywords of each of the related applications Fig.
14 shows another example of the step of determining representative keywords representing the target application among the keywords of the target application based on the weight of the keywords of the target application and the weight of the duplicate keywords of each of the related applications Fig.
FIG. 15 is a flowchart illustrating a method for determining representative keywords representative of the target application among the keywords of the target application based on the weight of the keywords of the target application and the weight of duplicate keywords of each of the related applications Fig.
FIG. 16 is a flowchart illustrating a process of determining representative keywords representing the target application among the keywords of the target application based on the weight of the keywords of the target application and the weight of duplicate keywords of each of the related applications, Fig.
17 is a diagram showing another example of the first database included in the representative keyword determination system of the application of FIG.

For the embodiments of the invention disclosed herein, specific structural and functional descriptions are set forth for the purpose of describing an embodiment of the invention only, and it is to be understood that the embodiments of the invention may be practiced in various forms, The present invention should not be construed as limited to the embodiments described in Figs.

The present invention is capable of various modifications and various forms, and specific embodiments are illustrated in the drawings and described in detail in the text. It is to be understood, however, that the invention is not intended to be limited to the particular forms disclosed, but on the contrary, is intended to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms may be used for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between. Other expressions that describe the relationship between components, such as "between" and "between" or "neighboring to" and "directly adjacent to" should be interpreted as well.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprise", "having", and the like are intended to specify the presence of stated features, integers, steps, operations, elements, components, or combinations thereof, , Steps, operations, components, parts, or combinations thereof, as a matter of principle.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries should be construed as meaning consistent with meaning in the context of the relevant art and are not to be construed as ideal or overly formal in meaning unless expressly defined in the present application .

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The same reference numerals are used for the same constituent elements in the drawings and redundant explanations for the same constituent elements are omitted.

1 is a diagram illustrating a representative keyword determination system of an application according to an exemplary embodiment of the present invention.

Referring to FIG. 1, a representative keyword determination system 10 of an application includes an explanation data collection unit 100, a keyword extraction unit 200, a first database DB1 300, an association application determination unit 400, And a keyword analysis unit 500.

The application market 20, which generally provides applications, provides explanatory materials (DS) describing the features, functions, etc. of the applications.

The representative keyword determination system 10 of the application crawls description data DS of each of a plurality of applications from the application market 20 via wired or wireless communication and stores description data DS of each of the plurality of applications, And determines representative keywords (REP_KWs) representative of each of the plurality of applications.

At this time, as will be described later, the representative keyword determination system 10 of the application acquires a list of associated applications having a high degree of association with each of the plurality of applications for each of the plurality of applications, The representative keywords (REP_KWs) representative of each of the plurality of applications are determined by analyzing not only the description data (DS) of each of the applications but also the description data (DS) of the associated applications for each of the plurality of applications .

Therefore, the representative keyword determination system 10 of the application according to the embodiments of the present invention can accurately determine the representative keywords (REP_KWs) that effectively show the characteristics of the application.

2 is a flowchart illustrating a representative keyword determination method of an application according to an embodiment of the present invention.

The representative keyword determination method of the application shown in FIG. 2 can be performed through the representative keyword determination system 10 of the application of FIG.

Hereinafter, a representative keyword determination method of an application performed through the representative keyword determination system 10 of the application will be described in detail with reference to FIGS. 1 and 2. FIG.

The explanation data collecting unit 100 can crawl description data DS of each of a plurality of applications from the application market 20 through wired / wireless communication (step S100).

For example, the description data collection unit 100 may sequentially search for and download explanatory data (DS) of all applications registered in the application market 20.

The description data collection unit 100 may provide the description data DS of each of the plurality of crawled applications to the keyword extraction unit 200. [

The keyword extracting unit 200 determines the specific gravity GV of the keywords KW from the keywords KW and the description data DS included in the description data DS for each of the plurality of applications May be stored in the first database 300 (step S200).

FIG. 3 is a flowchart illustrating an example of a step S200 of determining the weight of the keywords in the description data and the keywords in the description data for each of the plurality of applications shown in FIG. 2, and storing the determined weight in the first database .

The keyword extracting unit 200 extracts keywords (KW) from the keywords (KW) of each of the plurality of applications and descriptive data (DS) of the corresponding application by performing the operations shown in FIG. 3 for each of the plurality of applications (GW) occupied by each of the first and second servers (KW) may be determined and stored in the first database (300).

3, the keyword extracting unit 200 performs a morphological analysis on the description data DS for each of the plurality of applications to extract a keyword included in the description data DS (KW) (step S210).

In one embodiment, the keyword extractor 200 extracts all the morphemes included in the description data DS by performing a morphological analysis on the description data DS, and then describes the morphemes that are nouns among the identified morphemes (KW) of the data (DS).

However, the present invention is not limited thereto, and the keyword extracting unit 200 may extract keywords KW included in the description data DS using various types of morphological analysis algorithms.

Thereafter, the keyword extracting unit 200 may determine the number of repetitions of each of the keywords KW in the description data DS from which the keywords KW are extracted (step S220).

For example, the keyword extracting unit 200 may count the number of the keywords KW included in the description data DS and determine the number of repetitions of each of the keywords KW.

The keyword extracting unit 200 may then determine a value obtained by dividing the repetition count of each of the keywords KW by the total number of the repetitions of the keywords KW as the specific gravity GV of each of the keywords KW (Step S230).

For example, when the first keyword is repeated five times, the second keyword is repeated three times, and the third keyword is repeated twice in the explanation data DS, the keyword extracting unit 200 extracts the first keyword The specific gravity GV may be determined to be 0.5, the specific gravity GV of the second keyword may be determined to be 0.3, and the specific gravity GV of the third keyword may be determined to be 0.2.

The keyword extracting unit 200 associates the keywords GWs of the keywords KW and KW with the corresponding application name APP_N for each of the plurality of applications, (Step S240).

4 is a diagram showing an example of a first database included in the representative keyword determination system of the application of FIG.

Referring to FIG. 4, the first database 300a may include an application field APP_F, a keyword field KW_F, and a weight field GV_F.

The keyword extracting unit 200 stores the application name APP_N in the application field APP_F and stores the corresponding application keywords KW in the keyword field KW_F for each of the plurality of applications , And stores the specific gravity GVs of the corresponding application keywords KW in the gravity field GV_F.

4, the keywords KW of the application corresponding to APP1 are determined to be "conversation", "friend", and "attachment", and "conversation", "friend", and " Appendix "are determined to be" 0.40 ", "0.35 ", and" 0.25 ", and the keywords KW of the application corresponding to APP2 are determined as "pregnancy ","Quot;, " pregnancy ", "diary ", and" nutrition "in the application corresponding to APP2 are determined as" 0.10 ".

1 and 2, the associated application determination unit 400 can select a target application among the plurality of applications (step S300).

For example, as shown in FIG. 1, the associated application determination unit 400 selects one of application name (APP_N) stored in the first database 300 as a target application name (T_APP_N) Can be released.

Thereafter, the associated application determination unit 400 may obtain a list (CORR_APPs) of related applications having a high degree of association with the target application among the plurality of applications (step S400).

In one embodiment, the representative keyword determination system 10 of an application may previously store lists of associated applications for each of the plurality of applications.

In this case, the associated application determination unit 400 extracts, from lists of related applications for each of the plurality of applications stored in advance in the representative keyword determination system 10 of the application, a list of related applications corresponding to the target application (CORR_APPs (CORR_APPs) of the associated applications for the target application (step S400).

5 is a diagram showing an example of a representative keyword determination system of the application shown in FIG.

5, the representative keyword determination system 10 of the application shown in FIG. 5 may further include a second database (DB2) 600 in the representative keyword determination system 10 of the application shown in FIG. 1 have.

The second database 600 may store the list of related applications for each of the plurality of applications in advance.

In this case, the associated application determination unit 400 may read a list (CORR_APPs) of associated applications corresponding to the target application from the second database 600. [

6 is a diagram showing an example of a second database included in the representative keyword determination system of the application of FIG.

Referring to FIG. 6, the second database 600 may include an application field APP_F, an association application field CORR_APP_F, and an association index field RI_F.

The application field (APP_F) stores the name (APP_N) of each of the plurality of applications, the associated application field (CORR_APP_F) stores a list of associated applications having a high degree of association with the corresponding application, ) May store association indices indicating the degree of association between the corresponding application and the associated applications.

In one embodiment, the association index has a relatively large value as the degree of association between the corresponding application and the associated application is high, and the association index has a relatively small value as the association degree between the corresponding application and the associated application is low Value. ≪ / RTI >

6 illustratively shows that the associated applications of the application corresponding to APP1 are APP3, APP4 and APP7, and the association indices between APP1 and APP3, APP4 and APP7 are A1, A2, and A3, respectively, The associated applications of the application are APP4, APP5, APP8, and APP9, and the association indices between APP2 and APP4, APP5, APP8, and APP9 are A4, A5, A6, and A7, respectively. Here, A1, A2, A3, A4, A5, A6, and A7 represent real numbers.

In this case, the associated application determination unit 400 receives, from the second database 600, a list (CORR_APPs) of associated applications corresponding to the target application and association indexes RIs) can be read.

In another embodiment, the associated application determination unit 400 analyzes installation patterns of applications installed in a plurality of mobile terminals used by various users, and determines a list (CORR_APPs) of related applications having a high degree of association with the target application (Step S400).

7 is a block diagram showing an example of an associated application determination unit included in the representative keyword determination system of the application shown in FIG.

7, the associated application determination unit 400 includes a data receiving unit 410, a third database (DB3) 420, a ratio calculation unit 430, an association analysis unit 440, (450).

FIG. 8 is a flowchart illustrating an example of acquiring a list of associated applications having a high degree of association with a target application (S400) among the plurality of applications of FIG.

The step S400 of acquiring a list of related applications having a high degree of association with the target application among the plurality of applications shown in FIG. 8 may be performed through the associated application determination unit 400 shown in FIG.

7 and 8, the associated application determination unit 400 analyzes an installation pattern of applications installed in a plurality of mobile terminals 30 used by various users and determines an associated application having a high degree of association with the target application Lt; RTI ID = 0.0 > (CORR_APPs). ≪ / RTI >

Each of the plurality of mobile terminals 30 may internally store a unique identification ID.

In one embodiment, the identification ID may be an advertisement ID (ADID) uniquely assigned to each mobile terminal for the purpose of utilizing the advertisement.

Each of the plurality of mobile terminals 30 can download various types of applications from the application market 20, install them, and execute the installed applications.

Each of the plurality of mobile terminals 30 associates the installation log data I_LOG including the identification ID of each of the plurality of mobile terminals 30 and a list of all applications installed in each of the plurality of mobile terminals 30 To the application decision unit 400. [

In one embodiment, at least one of the applications installed in each of the plurality of mobile terminals 30 may include an install log data transfer module.

For example, applications that are prearranged among the plurality of applications registered in the application market 20 may include the installation log data transmission module in advance.

When the mobile terminal 30 downloads and installs the prearranged application from the application market 20, the installation log data transfer module may be installed together with the mobile terminal 30. [

The install log data transfer module is automatically executed in the mobile terminal 30 every time when the committed application is executed in the mobile terminal 30 and the identification log of the mobile terminal 30 is transmitted to the mobile terminal 30 And may transmit the installation log data I_LOG including the list of the installed applications to the associated application determination unit 400.

9 is a diagram illustrating an example of installation log data transmitted from each of a plurality of mobile terminals shown in FIG. 7 to an associated application determination unit.

9, the installation log data I_LOG includes semi-structured data including an identification ID (ADID) of the mobile terminal 30 and a list (APP_LIST) of applications installed in the mobile terminal 30, Lt; / RTI >

According to the embodiment, the installation log data I_LOG may further include a transmission date, an Internet Protocol (IP) address of the mobile terminal 30, and a model name of the mobile terminal 30. [

In FIG. 7, each of the plurality of mobile terminals 30 is shown as a smart phone, but the present invention is not limited thereto. Each of the plurality of mobile terminals 30 may be any portable mobile terminal such as a tablet computer, a mobile phone, a personal digital assistant (PDA), and the like.

7 and 8, the data receiving unit 410 included in the associated application determination unit 400 receives the identification ID (ADID) of each of the plurality of mobile terminals 30 from the plurality of mobile terminals 30, The installation log data I_LOG including a list of applications (APP_LIST) installed in each of the plurality of mobile terminals 30 may be received and stored in the third database 420 (step S410).

10 is a diagram showing an example of a third database included in the associated application determination unit of FIG.

Referring to FIG. 10, the third database 420 may include an identification ID field (ADID_F) and an application list field (APP_LIST_F).

When the data receiving unit 410 receives the installation log data I_LOG from each of the plurality of mobile terminals 30, the identification ID ADID included in the installation log data I_LOG is the identification of the third database 420 And the list APP_LIST of the applications included in the installation log data I_LOG may be stored in the application list field APP_LIST_F of the third database 420. [

10 shows an example in which applications such as APP1, APP3 and APP5 are installed in the mobile terminal 30 in which the identification ID ADID corresponds to ADID1 and the identification ID ADID is stored in the mobile terminal 30 corresponding to the ADID2. Indicates that applications such as APP2, APP3, and APP6 are installed.

7 and 8, the ratio calculation unit 430 included in the associated application determination unit 400 can select a target application among the plurality of applications stored in the first database 300 (step S300) .

For example, as shown in FIG. 1, the ratio calculation unit 430 selects one of application name (APP_N) stored in the first database 300 and reads it as the target application name (T_APP_N) .

Also, the ratio calculation unit 430 may determine the first to k-th applications to analyze the degree of association with the target application. Here, k represents a positive integer.

In one embodiment, the ratio calculation unit 430 may determine the first to k-th applications other than the target application among the plurality of applications stored in the first database 300. [

In another embodiment, the ratio calculation unit 430 may determine the first to k-th applications other than the target application among all the applications stored in the third database 420.

The ratio calculator 430 then calculates the ratio of the first to k-th (k-1) th mobile terminals 30 among the plurality of mobile terminals 30 based on the installation log data I_LOG of the plurality of mobile terminals 30 stored in the third database 420 It is possible to determine the first to kth overall ratios T_RATE_1 to T_RATE_k corresponding to the proportion of the mobile terminals 30 in which each of the applications is installed (step S420).

11 is a flowchart showing an example of a step S420 of determining the first to k-th overall ratios corresponding to the ratios of the mobile terminals in which the first to k-th applications are installed among the plurality of mobile terminals in Fig. 8 .

11, the ratio calculation unit 430 calculates the ratio of the mobile terminal 30 corresponding to the plurality of mobile terminals 30 based on the installation log data I_LOG of the plurality of mobile terminals 30 stored in the third database 420 The total number of identification IDs (ADID) can be determined (step S421).

As described above, since the identification ID (ADID) is an identification value uniquely assigned to mobile terminals, the total number may correspond to the total number of the plurality of mobile terminals 30.

The ratio calculation unit 430 calculates the ratio of the mobile terminal 30 to the mobile terminal 30 based on the installation log data I_LOG of the plurality of mobile terminals 30 stored in the third database 420 The number of identification IDs (ADID) can be determined as the j-th installation number (step S423). Here, j represents any positive integer equal to or greater than 1 and equal to or less than k.

Therefore, the number of the j-th installation may correspond to the number of the mobile terminals 30 in which the j-th application is installed among the plurality of mobile terminals 30. [

Then, the ratio calculation unit 430 may determine a j-th overall ratio (T_RATE_j) by dividing the number of the j-th installation by the total number (step S425).

Therefore, the j-th overall ratio T_RATE_j may represent the proportion of the mobile terminals 30 in which the j-th application is installed among the plurality of mobile terminals 30.

The ratio calculation unit 430 repeatedly performs the above-described steps S423 and S425 for each of the first to k-th applications, thereby calculating the ratio of the first to k-th applications among the plurality of mobile terminals 30 The first to kth total ratios T_RATE_1 to T_RATE_k corresponding to the ratio of the mobile terminals 30 installed can be determined.

Referring to FIGS. 7 and 8 again, the ratio calculation unit 430 calculates the ratio of the mobile terminal 30 installed with the target application based on the installation log data I_LOG of the plurality of mobile terminals 30 stored in the third database 420 (I_RATE_1 to I_RATE_k) corresponding to the ratio of the mobile terminals 30 installed with the first to k-th applications among the first to k-th applications 30 (step S430).

12 shows an example of a step (S430) of determining first to kth intersection ratios corresponding to ratios of mobile terminals in which each of the first to k-th applications among the mobile terminals in which the target application of Fig. 8 is installed It is a flowchart.

12, the ratio calculation unit 430 calculates the ratio of the mobile terminals 30 installed with the target application based on the installation log data I_LOG of the plurality of mobile terminals 30 stored in the third database 420, The number of identification IDs (ADID) corresponding to the target ID number (step S431).

Therefore, the target installation number may correspond to the number of the mobile terminals 30 in which the target application is installed among the plurality of mobile terminals 30.

The ratio calculation unit 430 may calculate the ratio of the mobile terminal 30 installed in the third database 420 based on the installation log data I_LOG of the plurality of mobile terminals 30 The number of identification IDs (ADID) corresponding to the number of identification IDs (ADID) 30 (step S433).

Therefore, the number of the j-th intersection may correspond to the number of the mobile terminals 30 installed together with the target application and the j-th application among the plurality of mobile terminals 30. [

Then, the ratio calculation unit 430 may determine the j-th intersection ratio I_RATE_j as a value obtained by dividing the j-th intersection number by the target installation number (step S435).

Therefore, the j-th intersection ratio I_RATE_j may indicate the ratio of the mobile terminals 30 in which the j-th application is installed among the mobile terminals 30 in which the target application is installed.

The ratio calculation unit 430 repeatedly performs the above-described steps S433 and S435 for each of the first to k-th applications to calculate the ratio of the first to k-th applications among the mobile terminals 30, The first to kth intersection ratios I_RATE_1 to I_RATE_k corresponding to the ratio of the mobile terminals 30 in which each of the applications is installed.

The ratio calculator 430 may provide the first to kth total ratios T_RATE_1 to T_RATE_k and the first to kth intersection ratios I_RATE_1 to I_RATE_k to the association analyzer 440.

7 and 8, the association analysis unit 440 included in the association application determination unit 400 determines whether the first to kth total ratios T_RATE_1 to T_RATE_k and the first to kth intersection ratios I_RATE_1 to I_RATE_k), the first to k-th association exponents RI_1 to RI_k indicating the degree of association between the target application and each of the first to k-th applications can be determined (step S440).

As described above, the j-th overall ratio T_RATE_j indicates the proportion of the mobile terminals 30 in which the j-th application is installed among the plurality of mobile terminals 30, and the j-th intersection ratio I_RATE_j indicates the ratio (I_RATE_j) is greater than the j-th total rate (T_RATE_j), the target application and the target application are not included in the mobile terminal 30 when the j-th intersection ratio I_RATE_j is greater than the j- There may be an inverse correlation between the target application and the j-th application if there is a correlation between the j-th application and the j-th intersection ratio I_RATE_j is smaller than the total ratio T_RATE_j have.

Here, the presence of correlation between the target application and the j-th application means that a ratio of the installed rate of the j-th application among the mobile terminals 30 in which the target application is installed is smaller than an average installation ratio of the j-th application And the presence of an inverse correlation between the target application and the j-th application means that the average rate of installation of the j-th application is higher than the average installation rate of the j-th application, It means that the rate of installing the j-th application is smaller.

In one embodiment, when there is a correlation between the target application and the j-th application, the association analyzing unit 440 determines that the j-th association index RI_j has a positive value, (RI_j), and when the inverse correlation exists between the target application and the j-th application, the j-th relevance index (RI_j) is determined such that the j-th association index (RI_j) has a negative value .

Specifically, the association analyzing unit 440 determines whether the j-th association index RI_j has a positive value and a relatively large absolute value as the correlation between the target application and the j-th application is high. j association index RI_j and determines that the j-th association index RI_j has a negative value and a relatively large absolute value as the inverse correlation between the target application and the j-th application increases. The j-th association index RI_j can be determined.

Hereinafter, the association analyzing unit 440 determines whether or not the target application and the first to k-th (k-1) th to k-th (k) based on the first to kth overall ratios T_RATE_1 to T_RATE_k and the first to kth intersection ratios I_RATE_1 to I_RATE_k. Various embodiments for determining the first to k < th > k association indices RI_1 to RI_k indicating the degree of association between each of the applications will be described.

In one embodiment, the association analyzing unit 440 calculates the correlation ratios (I_RATE_1 to I_RATE_k) based on the differences between the first to kth total ratios T_RATE_1 to T_RATE_k and the first to kth intersection ratios I_RATE_1 to I_RATE_k, 1 to k < th > association indexes RI_1 to RI_k.

For example, the association analyzer 440 may determine the first through k-th association exponents RI_1 through RI_k based on the following Equation (1).

[Equation 1]

J denotes a positive integer equal to or less than k, RI_j denotes the j-th association index, I_RATE_j denotes the j-th intersection ratio, and T_RATE_j denotes the j-th total ratio.

When the association analyzing unit 440 determines the first through k-th associated indexes RI_1 through RI_k based on Equation (1), the j-th relevance index RI_j may be determined based on the j- (30) having the j-th application and the ratio of the mobile terminals (30) having the j-th application installed among the plurality of mobile terminals (30) The j-th association index RI_j may objectively indicate a correlation or an inverse correlation between the target application and the j-th application.

As described above, the associated application determination unit 400 according to the embodiments of the present invention can determine whether or not a plurality of mobile terminals 30 have a plurality of mobile terminals 30 without considering the subjective preference evaluation of the first to k- The first to kth association indices RI_1 to RI_k are estimated by estimating a degree of association between the target application and each of the first to k-th applications based on a list (APP_LIST) of applications installed in the mobile terminals 30, Can be determined.

Therefore, the associated application determination unit 400 can effectively improve the objectivity and accuracy of the association analysis between applications.

In another embodiment, the association analyzing unit 440 may calculate the correlation ratios I_RATE_1 to I_RATE_k based on the ratios of the first to kth intersection ratios I_RATE_1 to I_RATE_k for the first to kth overall ratios T_RATE_1 to T_RATE_k, 1 to k < th > association indexes RI_1 to RI_k.

For example, the association analyzing unit 440 may determine the first to k-th association exponents RI_1 to RI_k based on the following Equation (2).

&Quot; (2) "

J denotes a positive integer equal to or less than k, RI_j denotes the j-th association index, I_RATE_j denotes the j-th intersection ratio, and T_RATE_j denotes the j-th total ratio.

The relevance analyzing unit 440 determines the first to k-th related indexes RI_1 to RI_k based on the above equation (2), the j-th relevance index RI_j is the j-th total ratio T_RATE_j, The j-th relevance index RI_j is determined based on the ratio between the first j-th intersection ratio I_RATE_j and the j-th intersection ratio I_RATE_j, even when the jth total ratio T_RATE_j and the j- Correlation and inverse correlation between the j-th application and the j-th application.

Also, when the average installation rate of the target application is relatively small, the size of the jth total ratio (T_RATE_j) and the jth intersection ratio (I_RATE_j) may be relatively small. In this case also, when the association analyzing unit 440 determines the first to k-th related indexes RI_1 to RI_k based on the above-mentioned Equation (2), the j-th association index RI_j is a Is determined based on the ratio between the ratio T_RATE_j and the j-th intersection ratio I_RATE_j, the associated application determination unit 400 according to the embodiments of the present invention effectively improves the accuracy of the association analysis between applications .

In another embodiment, the association analyzing unit 440 calculates the difference between each of the first to kth total ratios T_RATE_1 to T_RATE_k and the first to kth intersection ratios I_RATE_1 to I_RATE_k, The first to k-th associative exponents RI_1 to RI_k may be determined based on the ratio of each of the first to k-th intersection ratios I_RATE_1 to I_RATE_k to the kth total ratios T_RATE_1 to T_RATE_k, respectively .

For example, the association analyzer 440 may determine the first to k-th association exponents RI_1 to RI_k based on the following Equation (3).

&Quot; (3) "

J represents a positive integer equal to or less than k, RI_j represents the j-th association index, I_RATE_j represents the j-th intersection ratio, T_RATE_j represents the j-th total ratio, ABS A_DIFF_j represents a j-th absolute ratio difference, and R_DIFF_j represents a j-th relative ratio difference.

When comparing the above-mentioned Equations (1), (2) and (3), the j-th absolute ratio difference A_DIFF_j can be calculated by using the j-th association index RI_j And the j-th relative ratio difference R_DIFF_j is equal to the j-th association index RI_j according to Equation (2). Therefore, redundant description of the j-th absolute rate difference A_DIFF_j and the j-th relative rate difference R_DIFF_j will be omitted.

Referring to Equation (3), since the j-th absolute rate difference A_DIFF_j, the j-th relative rate difference R_DIFF_j, and the j-th related index RI_j have the same sign, ) Determines the first to k-th related indexes RI_1 to RI_k based on the above-described Equation (3), the j-th association index RI_j is a sum of the j-th absolute rate difference A_DIFF_j and the j- May correspond to the geometric mean of the difference R_DIFF_j.

Therefore, the j-th relevance index RI_j according to Equation (3) is the sum of the advantages of the association analysis between the target application and the j-th application in the j-th absolute rate difference A_DIFF_j and the j- ) May have all of the advantages of analyzing the association between the target application and the j-th application.

Therefore, the associated application determination unit 400 according to the embodiments of the present invention can more effectively improve the objectivity and accuracy of the association analysis between applications.

In another embodiment, the association analyzing unit 440 calculates a difference between each of the first to kth total ratios T_RATE_1 to T_RATE_k and each of the first to kth intersection ratios I_RATE_1 to I_RATE_k, Based on the ratio of each of the first through k-th intersection ratios I_RATE_1 through I_RATE_k for each of the k-th overall ratios T_RATE_1 through T_RATE_k, and the weights having a value larger than 0 and smaller than 1, The association indexes RI_1 to RI_k can be determined.

For example, the association analyzer 440 may determine the first through k-th association exponents RI_1 through RI_k based on Equation (4) below.

&Quot; (4) "

J denotes a positive integer equal to or less than k, RI_j denotes the j-th association index, I_RATE_j denotes the j-th intersection ratio, T_RATE_j denotes the j-th overall ratio, wt denotes the weight, ABS () represents an absolute value operation, A_DIFF_j represents a j-th absolute ratio difference, and R_DIFF_j represents a j-th relative ratio difference.

In one embodiment, the association analyzer 440 may obtain a weight (wt) through an input from the outside.

In an alternative embodiment, the association analyzer 440 may internally obtain the weight (wt) by reading the previously stored weight (wt).

The j-th relevance index RI_j according to Equation (4) can be calculated by substituting the j-th absolute ratio difference A_DIFF_j applying the weight wt and the j- May correspond to the weighted geometric mean of the relative ratio difference R_DIFF_j.

Therefore, the j-th relevance index RI_j according to Equation (4) can be calculated by taking advantage of the degree of association analysis between the target application and the j-th application possessed by the j-th absolute rate difference A_DIFF_j, ) May have all of the advantages of analyzing the association between the target application and the j-th application.

The influence of the j-th absolute ratio difference A_DIFF_j on the j-th relevance index RI_j and the influence of the j-th relative ratio difference R_DIFF_j on the j-th relevance index RI_j by adjusting the values of the weights wT Can be adjusted.

Therefore, the associated application determination unit 400 according to the embodiments of the present invention can further improve the accuracy of the association analysis between applications by adjusting the value of the weight (wt) according to the characteristics of the target application.

Referring again to FIGS. 7 and 8, the associated application selection unit 450 included in the associated application determination unit 400 determines whether or not the target application and the first to k- k association indices RI_1 to RI_k to determine the applications having a relatively high association index among the first to k-th applications as the related applications for the target application (step S450) .

In one embodiment, the associated application selecting unit 450 may determine a predetermined number of applications as the related applications in order of the association indexes among the first to k-th applications.

In another embodiment, the associated application selecting unit 450 may determine, among the first to k-th applications, applications having an association index higher than a threshold value as the associated applications.

Then, the associated application selecting unit 450 transmits the list of related applications (CORR_APPs) for the target application and the association indexes (RIs) indicating the association between the target application and the associated applications to the keyword analyzing unit 500, As shown in FIG.

5-12, description will be given of various embodiments in which the associated application determination unit 400 acquires a list (CORR_APPs) of associated applications having a high degree of association with the target application among the plurality of applications However, the present invention is not limited to this. According to embodiments, the associated application determination unit 400 may obtain a list (CORR_APPs) of related applications having a high degree of association with the target application in various ways.

Referring again to FIGS. 1 and 2, the keyword analyzer 500 may receive a list (CORR_APPs) of related applications for the target application from the associated application determination unit 400.

In addition, according to embodiments, the keyword analyzing unit 500 may further receive association indexes (RIs) indicating the degree of association between the target application and the associated applications from the associated application determination unit 400. [

Meanwhile, the keyword analyzing unit 500 may read the specific gravity GVs of the keywords KW and keywords of the target application and the related applications from the first database 300 (step S500).

Thereafter, the keyword analyzing unit 500 compares the keywords KW of the target application with the keywords KW of the associated applications KW, (KW) as duplicate keywords, and determines keywords that are not duplicate keywords among the keywords KW of the target application as non-duplicated keywords (step S600).

For example, the keyword analyzing unit 500 determines the keywords included in the keywords KW of the target application and the keywords KW of at least one of the related applications as the duplicate keywords, The non-duplicate keywords among the keywords KW of the target application can be determined as the non-duplicated keywords.

The keyword analyzing unit 500 analyzes the keywords of the target application KW based on the weighted GVs of the target applications KW and the weighted GVs of the redundant keywords of the associated applications, KW) representative keywords (REP_KWs) representative of the target application (step S700).

FIG. 13 is a flowchart illustrating a method of determining representative keywords representing the target application among the keywords of the target application (S700) based on the weight of the keywords of the target application of FIG. 2 and the weight of duplicate keywords of each of the associated applications; As shown in Fig.

Referring to FIG. 13, the keyword analyzing unit 500 may calculate a weight (GV) of the duplicate keywords of the target application for each of the duplicated keywords among the keywords KW of the target application, The score of the duplicate keyword can be determined by applying the specific gravity GV of the duplicate keyword as a weight (step S710).

In one embodiment, the keyword analyzing unit 500 may calculate a value obtained by adding a sum of the weights GV of the duplicate keywords of the associated applications to the specific gravity GV of the duplicate keyword of the target application, .

For example, if the specific gravity (GV) of the first duplicate keyword of the target application is a1, the first to third associated applications include the first duplicate keyword, The specific gravity GV of the first duplicate keyword is b1, the specific gravity GV of the first duplicate keyword of the second associated application is b2, and the specific gravity GV of the first duplicate keyword of the third related application is b3, the keyword analyzing unit 500 can determine the score of the first duplicate keyword of the target application as (a1 + b1 + b2 + b3).

On the other hand, the keyword analyzing unit 500 may calculate the specific gravity (GV) of the non-overlapping keyword of the target application as the score of the non-overlapping keyword for each of the non-overlapping keywords among the keywords KW of the target application (Step S730).

For example, if the weight of the first non-overlapping keyword of the target application is c1 and the associated application including the first non-overlapping keyword does not exist, the keyword analyzing unit 500 determines that the target The score of the first non-overlapping keyword of the application can be determined as c1.

After determining the scores of the duplicate keywords and the non-overlapping keywords of the target application in the above-described manner, the keyword analyzer 500 searches the target application keywords (KW) (REP_KWs) representative of the application (step S740).

In one embodiment, the keyword analyzing unit 500 may determine a predetermined number of keywords as representative keywords REP_KWs in descending order of the scores KW of the target application.

In another embodiment, the keyword analyzing unit 500 may determine keywords having scores higher than the threshold score among the keywords KW of the target application as representative keywords REP_KWs.

As described above with reference to FIG. 13, the representative keyword determination system 10 of the application according to the embodiments of the present invention not only calculates the weight (GV) of the keywords KW of the target application, And representative keywords (REP_KWs) representative of the target application are determined in consideration of the respective weights (GV) of the respective redundant keywords.

Therefore, the representative keyword determination system 10 of the application and the representative keyword determination method of the application according to the embodiments of the present invention can accurately determine representative keywords (REP_KWs) that effectively characterize the application.

FIG. 14 is a flowchart illustrating a process of determining representative keywords representing the target application among the keywords of the target application (S700) based on the weight of the keywords of the target application of FIG. 2 and the weight of duplicate keywords of each of the related applications; Fig.

Referring to FIG. 14, the keyword analyzing unit 500 may calculate, for each of the duplicate keywords among the keywords KW of the target application, the specific gravity GV of the duplicate keyword of the target application, In step S720, the score of the duplicate keyword may be determined by applying the weight of the duplicate keyword GV and the association index RI of each of the related applications as weights.

In one embodiment, the keyword analyzing unit 500 may calculate the product of the weight of the duplicate keyword (GV) of each of the associated applications and the association index (RI) of each of the associated applications, (GV) may be determined as the score of the duplicate keyword.

For example, if the specific gravity (GV) of the first duplicate keyword of the target application is a1, the first to third associated applications include the first duplicate keyword, The specific weight (GV) of the first duplicate keyword is b1, the association index (RI) of the first associated application is r1, the specific gravity (GV) of the first duplicate keyword of the second associated application is b2, When the association index (RI) of the second associated application is r2, the specific gravity (GV) of the first duplicate keyword of the third associated application is b3, and the association index (RI) of the third associated application is r3, The analyzer 500 may determine the score of the first duplicate keyword in the target application as (a1 + b1 * r1 + b2 * r2 + b3 * r3).

On the other hand, the keyword analyzing unit 500 may calculate the specific gravity (GV) of the non-overlapping keyword of the target application as the score of the non-overlapping keyword for each of the non-overlapping keywords among the keywords KW of the target application (Step S730).

For example, if the weight of the first non-overlapping keyword of the target application is c1 and the associated application including the first non-overlapping keyword does not exist, the keyword analyzing unit 500 determines that the target The score of the first non-overlapping keyword of the application can be determined as c1.

After determining the scores of the duplicate keywords and the non-overlapping keywords of the target application in the above-described manner, the keyword analyzer 500 searches the target application keywords (KW) (REP_KWs) representative of the application (step S740).

As described above with reference to FIG. 14, the representative keyword determination system 10 of the application according to the embodiments of the present invention not only calculates the weight (GV) of the keywords KW of the target application, The representative keywords REP_KWs representative of the target application are determined in consideration of the weight of each of the redundant keywords GV and the association indexes RIs of the related applications.

Therefore, the representative keyword determination system 10 of the application and the representative keyword determination method of the application according to the embodiments of the present invention can more accurately determine representative keywords (REP_KWs) that effectively characterize the application.

FIG. 15 is a flowchart illustrating a process of determining representative keywords representing the target application among the keywords of the target application (S700) based on the weight of the keywords of the target application of FIG. 2 and the weight of duplicate keywords of each of the related applications; As shown in Fig.

Referring to FIG. 15, the keyword analyzing unit 500 analyzes the target application KV based on the weighted GVs of the keywords KW of the target application and the weighted GVs of the redundant keywords of each of the associated applications, The score of each of the keywords KW of the user can be determined (step S701).

In one embodiment, the keyword analysis unit 500 determines the score of each of the duplicate keywords of the target application and the score of each of the non-overlapping keywords of the target application in the manner described above with reference to FIG. 13 .

In another embodiment, the keyword analysis unit 500 determines the score of each of the duplicate keywords of the target application and the score of each of the non-overlapping keywords of the target application in the manner described above with reference to Fig. 14 .

Then, the keyword analyzing unit 500 analyzes the number of the keywords included in the target application KW based on the number of applications including the keywords overlapping each of the keywords KW of the target application among the plurality of applications stored in the first database 300 The correction coefficient of each of the keywords KW can be determined (step S750).

For example, if the number of applications having the same keyword as the keyword KW of the target application among the plurality of applications stored in the first database 300 is relatively large, The correction coefficient of the application keyword KW is determined to be a relatively small value and the number of applications including the keyword identical to the keyword KW of the target application among the plurality of applications stored in the first database 300 Is relatively small, the correction coefficient of the keyword (KW) of the target application can be determined to be a relatively large value.

Then, the keyword analyzing unit 500 compares the score of each of the keywords KW of the target application with the correction coefficient of each of the keywords KW of the target application to the keywords KW of the target application, (Step S770), and determines keywords having a relatively higher correction score among the keywords KW of the target application as representative keywords REP_KWs representative of the target application (step S780) ).

Generally, common words commonly used in DSs of various applications such as "authority "," settlement ", etc. can not effectively represent characteristics or functions of a specific application.

As described above with reference to Fig. 15, the representative keyword determination system 10 of the application according to the embodiments of the present invention can use the correction coefficient to determine description data DS of various applications such as " It is possible to effectively prevent general keywords from being determined as the representative keywords (REP_KWs) of the target application.

Therefore, the representative keyword determination system 10 of the application and the representative keyword determination method of the application according to the embodiments of the present invention can more accurately determine representative keywords (REP_KWs) that effectively characterize the application.

FIG. 16 is a flowchart illustrating a method of determining representative keywords representative of the target application among the keywords of the target application (S700) based on the weight of the keywords of the target application of FIG. 2 and the weight of duplicate keywords of each of the related applications; As shown in Fig.

Referring to FIG. 16, the keyword analyzing unit 500 analyzes the target application (GW) based on the GWs of the keywords KW of the target application and the GVs of the redundant keywords of the respective applications, The score of each of the keywords KW of the user can be determined (step S701).

In one embodiment, the keyword analysis unit 500 determines the score of each of the duplicate keywords of the target application and the score of each of the non-overlapping keywords of the target application in the manner described above with reference to FIG. 13 .

In another embodiment, the keyword analysis unit 500 determines the score of each of the duplicate keywords of the target application and the score of each of the non-overlapping keywords of the target application in the manner described above with reference to Fig. 14 .

In one embodiment, the description data DS of each of the plurality of applications crawled from the application market 20 may include category information of the plurality of applications.

In this case, the keyword extracting unit 200 may additionally store the category information of each of the plurality of applications in the first database 300 based on the description data DS of each of the plurality of applications.

17 is a diagram showing another example of the first database included in the representative keyword determination system of the application of FIG.

Referring to FIG. 17, the first database 300b may include an application field APP_F, a category field CATEGORY_F, a keyword field KW_F, and a specific gravity field GV_F.

The keyword extracting unit 200 stores the name (APP_N) of the application in the application field (APP_F) for each of the plurality of applications, stores the category of the corresponding application in the category field (CATEGORY_F) The keywords KW of the corresponding application KW can be stored in the keyword field KW_F and the specific gravity GV of the corresponding application keywords KW can be stored in the gravity field GV_F.

17, the category of the application corresponding to APP1 is "communication", the keywords KW of the application corresponding to APP1 are determined as "conversation", "friend", and "attachment" 0.35 " and "0.25 ", respectively, and the category of the application corresponding to APP2 is" birth " , The keywords KW of the application corresponding to APP2 are determined as "pregnancy", "fetish", "diary", and "nutrition", and "pregnancy", " , And the "specific gravity" (GV) of "nutrition" were determined as "0.45", "0.30", "0.15", and "0.10", respectively.

In this case, as shown in FIG. 16, the keyword extracting unit 200 extracts, from among the plurality of applications stored in the first database 300, an application including keywords overlapping each of the keywords KW of the target application (KW) of the target application based on the number of categories to which the target application belongs (step S760).

For example, when the number of categories to which the applications including the keyword (KW) of the target application belongs is relatively large among the plurality of applications stored in the first database 300 , Determines the correction coefficient of the keyword (KW) of the target application to be a relatively small value, and determines the correction coefficient of the keyword (KW) of the target application as a relatively small value, When the number of categories to which the applications belong is relatively small, the correction coefficient of the keyword (KW) of the target application can be determined to be a relatively large value.

Then, the keyword analyzing unit 500 compares the score of each of the keywords KW of the target application with the correction coefficient of each of the keywords KW of the target application to the keywords KW of the target application, (Step S770), and determines keywords having a relatively higher correction score among the keywords KW of the target application as representative keywords REP_KWs representative of the target application (step S780) ).

Generally, common words commonly used in description data (DS) of applications belonging to various categories of categories such as "authority" and "settlement" can not effectively represent the characteristics or functions of a specific application.

As described above with reference to Figs. 16 and 17, the representative keyword determination system 10 of the application according to the embodiments of the present invention may be configured to determine the category of the category, such as " It is possible to effectively prevent general keywords from being determined as the representative keywords (REP_KWs) of the target application by decreasing the scores of common keywords commonly used in the description data DS of the applications to which the application belongs.

Therefore, the representative keyword determination system 10 of the application and the representative keyword determination method of the application according to the embodiments of the present invention can more accurately determine representative keywords (REP_KWs) that effectively characterize the application.

The representative keyword determination system 10 of the application according to the embodiments of the present invention sequentially selects each of the plurality of applications stored in the first database 300 as the target application, The representative keywords (REP_KWs) of each of the plurality of applications can be determined by repeatedly performing the operations described above with reference to FIGS.

The representative keywords (REP_KWs) of each of the plurality of applications determined through the representative keyword determination system 10 of the application according to the embodiments of the present invention can be utilized in various fields such as searching for an application, have.

INDUSTRIAL APPLICABILITY The present invention can accurately determine representative keywords that effectively characterize an application, and thus can be used to improve the convenience of application search and to select an advertisement object of an application.

Although the preferred embodiments of the present invention have been disclosed for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, without departing from the scope and spirit of the invention as disclosed in the accompanying claims. It will be understood that the invention may be modified and varied without departing from the scope of the invention.

10: Application's representative keyword determination system
20: Application Market 30: Mobile Terminal
100: description data collecting unit 200: keyword extracting unit
300: First database 400: Associated application determining unit
410: Data receiving unit 420: Third database
430: ratio calculation unit 440: relevance analysis unit
450: Associated application selection unit 500: Keyword analysis unit
600: second database

Claims (20)

Crawling description data of each of a plurality of applications from an application market;
Determining, for each of the plurality of applications, the weight of the keywords in the description data and the keywords included in the description data and storing the determined weight in the first database;
Selecting a target application among the plurality of applications;
Obtaining a list of associated applications having a high degree of association with the target application among the plurality of applications;
Reading the weight of the keywords and the keywords of each of the target application and the associated applications from the first database;
Determining duplicated keywords as duplicate keywords among the keywords of the associated applications among the keywords of the target application; And
Determining representative keywords representative of the target application among the keywords of the target application based on the weight of the keywords of the target application and the weight of the duplicate keywords of each of the associated applications How to determine representative keywords. 2. The method of claim 1, wherein for each of the plurality of applications, determining the weight of the keywords from the keywords included in the description data and the description data and storing the determined weight in the first database comprises:
For each of the plurality of applications,
Performing morphological analysis on the explanatory data to extract the keywords included in the explanatory data;
Determining the number of repetitions of each of the keywords in the description data;
Determining a value obtained by dividing the repetition count of each of the keywords by the total of the repetition counts of the keywords as the weight of each of the keywords; And
And associating the weight of the keywords and the keywords with a name of a corresponding application and storing the keyword in the first database. The method of claim 1, wherein the step of obtaining a list of the associated applications having a high degree of association with the target application among the plurality of applications comprises:
Reading, for each of the plurality of applications, a list of the associated applications corresponding to the target application from a second database that stores in advance a list of the associated applications having a high degree of association with the corresponding application A method for determining representative keywords of an application. The method of claim 1, wherein the step of obtaining a list of the associated applications having a high degree of association with the target application among the plurality of applications comprises:
Receiving installation log data including an identification ID of each of the plurality of mobile terminals and a list of applications installed in each of the plurality of mobile terminals, and storing the received installation log data in a third database;
(K is a positive integer) applications among the plurality of mobile terminals based on the installation log data of the plurality of mobile terminals stored in the third database Determining first through kth total ratios;
K-th applications corresponding to the ratios of the mobile terminals in which the first to k-th applications are installed, among the mobile terminals in which the target application is installed, based on the installation log data of the plurality of mobile terminals stored in the third database k crossing ratios;
Determining first to k < th > k association indices indicative of a degree of association between the target application and each of the first to k-th applications based on the first to k-th overall ratios and the first to k- ; And
And determining, as the related applications, applications having a relatively high association index among the first to k-th applications. 5. The method according to claim 4, further comprising the step of, based on the installation log data of the plurality of mobile terminals stored in the third database, Wherein determining the first through k < RTI ID = 0.0 > k <
Determining a total number of the identification IDs corresponding to the plurality of mobile terminals based on the installation log data of the plurality of mobile terminals;
Determining a number of the identification IDs corresponding to mobile terminals provided with an application of j (j is an arbitrary positive integer equal to or greater than 1 k) based on the installation log data of the plurality of mobile terminals as a j-th installation number ; And
Determining a value obtained by dividing the number of the j-th installation by the total number as a j-th overall ratio. 5. The method according to claim 4, further comprising the steps of: determining, based on the installation log data of the plurality of mobile terminals stored in the third database, a ratio of mobile terminals installed with the first to k- The step of determining corresponding first through kth intersection ratios comprises:
Determining, as a target installation number, the number of the identification IDs corresponding to the mobile terminals on which the target application is installed based on the installation log data of the plurality of mobile terminals;
(J is a positive integer equal to or less than k) applications based on the installation log data of the plurality of mobile terminals, the number of the identification IDs corresponding to mobile terminals installed together with the target application and the j j as a number of intersections; And
Determining a value obtained by dividing the j-th intersection number by the target installed number as a j-th intersection ratio. 5. The method of claim 4, further comprising: determining whether the target application and the first to k-th applications are associated with each other based on the first through k < 0.0 > k < / RTI > association indexes,
And determining the first through k-th associated indexes based on a difference between each of the first through k-th overall ratios and each of the first through k-th crossing ratios. 5. The method of claim 4, further comprising: determining whether the target application and the first to k-th applications are associated with each other based on the first through k < 0.0 > k < / RTI > association indexes,
And determining the first through k k-related indexes based on a ratio of each of the first through k-th crossing ratios for each of the first through k-th overall ratios. 5. The method of claim 4, further comprising: determining whether the target application and the first to k-th applications are associated with each other based on the first through k < 0.0 > k < / RTI > association indexes,
The ratio of each of the first through kth total ratios and each of the first through kth intersection ratios to each of the first through kth total ratios, And determining the first through k < th > k-related indexes based on the first through k-th associated indexes. 5. The method of claim 4, further comprising: determining whether the target application and the first to k-th applications are associated with each other based on the first through k < 0.0 > k < / RTI > association indexes,
A ratio of each of the first through k-th total ratios and a difference between each of the first through k-th intersection ratios, a ratio of each of the first through k-th intersection ratios to each of the first through k- And determining the first through k-th associated indexes based on a weight having a value greater than zero and less than one. The method of claim 1, further comprising: determining the representative keywords representing the target application from among the keywords of the target application based on the weight of the keywords of the target application and the weight of the duplicate keywords of each of the associated applications Lt; / RTI >
For each of the duplicate keywords among the keywords of the target application, a weight of the duplicate keyword is determined by applying a weight of the duplicate keyword of each of the associated applications to a weight of the duplicate keyword of the target application step;
Determining a weight of the non-overlapping keyword of the target application as the score of the non-overlapping keyword, for each of the non-overlapping keywords that are not duplicate keywords among the keywords of the target application; And
And determining, as the representative keywords, keywords having a relatively high score among the keywords of the target application. 12. The method of claim 11, wherein for each of the duplicate keywords among the keywords in the target application, a weight of the duplicate keyword of each of the associated applications is applied as a weight to a weight of the duplicate keyword of the target application, The step of determining the score of the keyword comprises:
Determining a sum of the weights of the duplicate keywords of the associated applications and a weight of the duplicate keyword of the target application as a score of the duplicate keyword. The method of claim 1, wherein the step of obtaining a list of the associated applications having a high degree of association with the target application among the plurality of applications comprises:
Obtaining a list of the associated applications corresponding to applications having a relatively high degree of association with the target application among the plurality of applications and association indexes indicating association degrees between the target applications and the associated applications A method for determining representative keywords of an application. 14. The method of claim 13, further comprising determining the representative keywords representing the target application from the keywords of the target application based on the weight of the keywords in the target application and the weight of the duplicate keywords in each of the associated applications Lt; / RTI >
For each of the duplicate keywords among the keywords of the target application, a weight of the duplicate keyword of each of the associated applications and a weight of the association index of each of the associated applications in the weight of the duplicate keyword of the target application Determining a score of the duplicate keyword;
Determining a weight of the non-overlapping keyword of the target application as the score of the non-overlapping keyword, for each of the non-overlapping keywords that are not duplicate keywords among the keywords of the target application; And
And determining, as the representative keywords, keywords having a relatively high score among the keywords of the target application. 15. The method of claim 14, further comprising: for each of the duplicate keywords among the keywords in the target application, the weight of the duplicate keyword of each of the associated applications in the weight of the duplicate keyword in the target application, The step of determining the score of the duplicate keyword by applying the association index as a weight,
Determining a sum of the weight of the duplicate keyword of each of the related applications and the association index of each of the associated applications together with the weight of the duplicate keyword of the target application as the score of the duplicate keyword How to determine representative keywords. The method of claim 1, further comprising: determining the representative keywords representing the target application from among the keywords of the target application based on the weight of the keywords of the target application and the weight of the duplicate keywords of each of the associated applications Lt; / RTI >
Determining a score of each of the keywords in the target application based on the weight of the keywords in the target application and the weight of the duplicate keywords in each of the associated applications;
Determining a correction coefficient of each of the keywords of the target application based on the number of applications including a keyword overlapping each of the keywords of the target application among the plurality of applications;
Determining a value obtained by multiplying the score of each of the keywords of the target application by the correction coefficient of each of the keywords of the target application as a correction score of each of the keywords of the target application; And
And determining, as the representative keywords, keywords having a relatively high correction score among the keywords of the target application. 17. The method of claim 16, wherein determining the correction factor for each of the keywords of the target application based on a number of applications that include a keyword overlapping each of the keywords of the target application among the plurality of applications ,
Determining the correction coefficient of the keyword of the target application as a relatively small value when the number of applications including the same keyword as the keyword of the target application is relatively large among the plurality of applications; And
Determining the correction coefficient of the keyword of the target application to be a relatively large value when the number of applications including the same keyword as the keyword of the target application is relatively small among the plurality of applications, ≪ / RTI > The method according to claim 1,
Further comprising storing the category information of each of the plurality of applications in the first database based on the description data of each of the plurality of applications,
The step of determining the representative keywords representing the target application among the keywords of the target application based on the weight of the keywords of the target application and the weight of the redundant keywords of each of the related applications,
Determining a score of each of the keywords in the target application based on the weight of the keywords in the target application and the weight of the duplicate keywords in each of the associated applications;
Determining a correction coefficient of each of the keywords of the target application based on the number of categories to which applications including keywords overlapping each of the keywords of the target application among the plurality of applications belong;
Determining a value obtained by multiplying the score of each of the keywords of the target application by the correction coefficient of each of the keywords of the target application as a correction score of each of the keywords of the target application; And
And determining, as the representative keywords, keywords having a relatively high correction score among the keywords of the target application. 19. The method of claim 18, further comprising: determining the correction factor of each of the keywords of the target application based on the number of categories to which applications containing keywords overlapping each of the keywords of the target application among the plurality of applications belong. Lt; / RTI >
Determining the correction coefficient of the keyword of the target application to be a relatively small value when the number of categories to which the applications including the same keyword as the keyword of the target application belongs is relatively large among the plurality of applications; And
Determining the correction coefficient of the keyword of the target application to be a relatively large value when the number of categories to which the applications including the same keyword as the keyword of the target application belongs is relatively small among the plurality of applications, A method for determining representative keywords of an application to be included. Database;
A description data collection unit for crawling description data of each of a plurality of applications from an application market;
A keyword extracting unit for each of the plurality of applications to determine the specific weight of the keywords from the keywords included in the explanatory data and the explanatory data and store the determined weight in the database;
An associated application determining unit selecting a target application among the plurality of applications and acquiring a list of related applications having a high degree of association with the target application among the plurality of applications; And
Reads out the weight of the keywords and the keywords of each of the target application and the related applications from the database and determines keywords that are overlapped with the keywords of the related applications among the keywords of the target application as duplicate keywords And a keyword analysis unit for determining representative keywords representing the target application among the keywords of the target application based on the weight of the keywords of the target application and the weight of the redundant keywords of each of the related applications A representative keyword determination system of an application to be executed.

Images