KR20160040341A - Method of analyzing a stream of a business area and system for performing the same - Google Patents

Abstract

Disclosed is a method of analyzing a stream of a business area and a system for performing the same. The system of analyzing a stream of a business area comprises: a web document collection unit which collects a plurality of web documents; a morpheme analysis unit which analyzes a morpheme in the collected web documents; a morpheme classification-storing unit which classifies and stores the analyzed morphemes based on a region; and a determining unit for stream of the business area which determines the region of the business area based on the number of the morpheme stored based on the region. Thus, the stream of the business area is figured out by extracting information from a real time web document such as a social commerce, a concert, an SNS, a blog, a news, thereby the business area for the region may be analyzed based on the stream of time for region information and a region characteristic.

Description

[0001] METHOD OF ANALYZING A STREAM OF A BUSINESS AREA AND SYSTEM FOR PERFORMING THE SAME [0002]

Field of the Invention The present invention relates to a method and an apparatus for analyzing a regional commercial flow, and more particularly, to a method for analyzing a local commercial flow through a web analysis and a system for performing the same.

Recently, as the use of the Internet becomes more and more active, many people use social networks such as bulletin boards, blogs, community cafes, social network services such as Facebook and mini homepage We express our opinion through online media. Therefore, users often make decisions by referring to the information that other users post on the Internet when evaluating the value of certain information.

For example, there are opinions from various users, from product reviews to movie reviews, on web portals on the Internet, specialized websites for industry, and personal blogs. These users' opinions can be used by general users as references for purchasing products or viewing movies, and even when a marketing person or a stock trader wants to know the opinions of users about each product or company .

Korean Patent Laid-Open No. 10-2013-0030434 (Name: Web-based commercial document analysis service system and method published online) (Published on March 23, 2013) Korean Patent Laid-Open Publication No. 10-2012-0035600 (name: system and method for providing information on merchandise information, merchandise information providing apparatus, service apparatus and information providing terminal for merchandise information) (disclosed on Apr. 16, 2012) Korean Patent Laid-Open No. 10-2013-0019629 entitled " System and method for locating and analyzing commercial goods by types of real estate goods, and computer-readable recording medium " Korean Registered Patent No. 10-1078344 (Name: How to estimate the sales by industry in the trade) (Registered on October 25, 2011)

SUMMARY OF THE INVENTION Accordingly, the present invention has been made keeping in mind the above problems occurring in the prior art, and it is an object of the present invention to provide a method and apparatus for analyzing a local commercial web site through a web analysis, which extracts information from a real-time web document such as a social commerce, And to provide a method for analyzing local commercial flow to understand the flow.

It is another object of the present invention to provide a system for performing the method of analyzing local commercial flow.

In order to accomplish the object of the present invention, a method for analyzing a local merchandise flow according to an exemplary embodiment includes: collecting a plurality of web documents; Performing morpheme analysis on the collected web documents; Classifying and analyzing the analyzed morphemes by region; And determining a commercial area based on the number of morphemes stored on the area basis.

In one embodiment, the morpheme may include a point morpheme corresponding to the information registered in the point of interest (POI) DB and a position morpheme corresponding to the position information not registered in the point of interest DB.

In one embodiment, the step of classifying and storing the analyzed morphemes by region includes extracting the morpheme of interest or the morpheme in the analyzed morphemes; And classifying and storing each of the interest point morpheme and the position morpheme on an area basis.

In one embodiment, the step of discriminating the commercial area may discriminate the commercial area based on the number of the interest point morpres and the number of the position morpreses stored on the area basis.

In one embodiment, the step of discriminating the commercial area includes analyzing a commercial area corresponding to the area, analyzing the commercial area according to the area, analyzing the commercial area corresponding to the area and period, analyzing the commercial area according to the period, A commercial document analysis step for analyzing a commercial document, and a commercial document analysis step for analyzing a commercial article according to an interest point category.

In one embodiment, the web document may include a social commerce, a performance, an SNS, a blog, and a news.

According to another aspect of the present invention, there is provided a regional merchant flow analysis system comprising: a web document collection unit for collecting a plurality of web documents; A morphological analysis unit for analyzing morphemes in the collected web documents; A morpheme classifying and storing unit for classifying and analyzing the analyzed morphemes by region; And a merchandise flow discrimination unit for discriminating a merchandise area based on the number of morphemes stored on the area basis.

In one embodiment, the morpheme may include a point morpheme corresponding to the information registered in the point of interest (POI) DB and a position morpheme corresponding to the position information not registered in the point of interest DB.

In one embodiment, the morpheme classification-storing unit includes: a morpheme extracting unit for extracting the morpheme of interest and the morpheme corresponding to the position information in the analyzed morphemes; And a morpheme classifier for classifying each of the interest point morpheme and the position morpheme on an area basis.

In one embodiment, the local merchant flow analysis system includes an interest point morpheme store DB that stores the interest point morpheme classified by the morpheme classifier; And a location morpheme storage DB for storing location morphemes classified by the morpheme classification unit.

In one embodiment, the merchandise flow determination unit determines the number of points of interest morpresent stored in the interest point morpheme storage DB and the number of position morpheme stores stored in the location morpheme store DB by the regional reference, The area can be identified.

In one embodiment, the merchandise flow determination unit may include: a regional commercial analysis module for analyzing a commercial area corresponding to a region; a commercial analysis module for each period to analyze a commercial area corresponding to a region and a period; A commercial document analysis module for each type of web document analyzing the commercial document category, and a commercial document analysis module for each interest category for analyzing commercial articles by the interest point category.

According to the method of analyzing local commercial flow and the system for performing the same, information is extracted from real-time web documents such as social commerce, performance, SNS, blog, and news, and stored and analyzed, , Regional information or local characteristics can be analyzed over time to analyze commercial areas. .

FIG. 1 is a block diagram for explaining a local commercial flow analysis system according to an embodiment of the present invention.
FIG. 2 is a block diagram for explaining the commercial flow determiner shown in FIG. 1. FIG.
FIG. 3 is a flowchart illustrating a method of analyzing a local commercial floor flow according to an exemplary embodiment of the present invention.
FIG. 4 is a flowchart for explaining the step of discriminating the commercial area shown in FIG.
FIG. 5 is a flowchart for explaining the regional merchandise analysis steps shown in FIG.
FIG. 6 is a flow chart for explaining the commercial article analysis step according to the period shown in FIG.
FIG. 7 is a flow chart for explaining the merchandise analysis step for each web document type shown in FIG.
FIG. 8 is a flowchart for explaining a commercial area analysis step for each point of interest category shown in FIG.
9 is a graph for explaining the total number of interest point morphemes and location morphemes extracted by region and region in a web document.
10 is a graph for explaining an example of the total number of points of interest (POI) morphemes corresponding to each area.
11 is a graph for explaining an example of the density of the point of interest (POI) per 2D unit area of each area.
FIG. 12 is a graph for explaining an example of the density of the point of interest (POI) per 3D unit area of the area.
13 is a graph for explaining the simple commercial exponent index. In particular, a graph corresponding to Table 6 is shown.
FIG. 14 is a graph for explaining the 2D commercial index reflecting the 2D area.
FIG. 15 is a graph for explaining the 3D premium index reflecting the 3D area.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in more detail with reference to the accompanying drawings. 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 should be understood, however, that the invention is not intended to be limited to the particular forms disclosed, but includes 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 are used only 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. The singular expressions include plural expressions unless the context clearly dictates otherwise.

In this application, the terms "comprises", "having", and the like are used to specify that a feature, a number, a step, an operation, an element, a part or a combination thereof is described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, components, parts, or combinations thereof.

Also, 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 are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted as either ideal or overly formal in the sense of the present application Do not.

Prior to the description, the web document referred to in the present invention includes texts posted by netizens on the online medium, comments added to the texts, attachments or images, and the like, such as affirmation or negation of netizen Include comments.

The online medium is a community between other users, such as web sites (newspaper sites, broadcast sites, specialty store sites), bulletin boards, blogs, cafes, sites that support social network services, etc. where web documents can be generated by netizens Means a possible channel.

FIG. 1 is a block diagram for explaining a local commercial flow analysis system according to an embodiment of the present invention.

1, the local merchant flow analysis system 100 includes a web document collection unit 110, a web document storage DB 120, a morphological analysis unit 130, a morphological classification-storage unit 140, A storage DB 150, a location morpheme storage DB 160, and a merchant flow determination unit 170. 1, the local merchandise flow analysis system 100 includes the web document collection unit 110, the web document storage DB 120, the morpheme analysis unit 130, the morpheme classification-storage unit 140, A point of interest (POI) morpheme storage DB 150, the location morpheme storage DB 160, and the merchandise flow discrimination unit 170. However, this is logically divided for convenience of explanation It is not a hardware distinction.

Point of interest (POI) data is data that displays an electronic digital map with coordinates of major facilities, stations, airports, terminals, hotels, and the like. It can be classified into search data used for destination search and background data only displayed on the desktop. If the user knows the information about the destination, that is, the address or telephone number of the destination or the exact name, the user can directly search for the destination by directly inputting the point of interest data.

In this embodiment, the point of interest morpheme refers to the text (i. E., Character) corresponding to the POI data, and the location morpheme is text that is not registered in the point of interest morpheme but refers to a location commonly used by the public.

The web document collection unit 110 collects a plurality of web documents and stores the collected web documents in the web document storage DB 120. The web document may include social commerce, performance, SNS, blog, news, and the like.

The morpheme analysis unit 130 analyzes the morphemes in the web documents stored in the web document DB 120.

The morpheme classifying and storing unit 140 includes a morpheme extracting unit 142 and a morpheme sorter 144. The morpheme sorter storing unit 140 classifies morphemes analyzed by the morpheme analyzing unit 130 into regions and stores them in the interest point morpheme storage DB 150) or the location morpheme storage DB (160).

The morpheme extracting unit 142 extracts a morpheme corresponding to the interest point morpheme or the position information corresponding to the interest point information in the analyzed morphemes and provides the extracted morpheme to the morpheme sorter 144.

The morpheme classifier 144 classifies the morphemes of interest or position morphemes provided by the morpheme extractor 142 on the basis of regions and stores them in the interest point morpheme storage DB 150 or the location morpheme storage DB 160 .

The interest point morpheme storage DB 150 stores the interest point morpheme classified by the morpheme classifying unit 144 for each region.

The location morpheme storage DB 160 stores location morphemes classified by the morpheme classification unit 144 for each region.

The merchandise flow determiner 170 determines the merchant area based on the number of morphemes stored on an area basis, that is, the number of morpheme points of interest and the number of morphemes.

As described above, according to the present embodiment, attention is paid to the fact that regional information and regional characteristics are analyzed as a flow of time and regional trade analysis is required. Thus, it is possible to analyze commercials such as social commerce, performances, SNS, We collect real-time web documents, extract location information and points of interest (POIs) from the collected web documents, classify and analyze them on a regional basis, and then identify popular areas based on the number of times mentioned. In particular, it is possible to determine which category is popular depending on the type of the extracted web document and the POI category.

FIG. 2 is a block diagram for explaining the commercial flow determiner shown in FIG. 1. FIG.

Referring to FIG. 2, the merchandise flow discrimination unit 170 includes a regional merchandise analysis module 172, a merchandise trade analysis module 174, a merchandise trade analysis module 176 according to a web document type, a merchandise trade analysis module 178, And an interface module 179, and determines a commercial area based on the number of morphemes stored on an area basis. 2, the merchandise flow determiner 170 receives the merchandise trade analysis module 172, the merchandise trade analysis module 174, the web site category analysis module 176, 178 and the interface module 179. However, the interface module 179 is logically divided for the sake of convenience of description, but is not limited to hardware.

The regional commercial value analysis module 172 provides the terminal 300 with information for area selection through the interface module 179. [ Providing the information for the area selection to the terminal 300 may be omitted. For example, if the regional commercial analysis is programmed to be performed in real time or periodically, providing the information for the area selection to the terminal 300 may be omitted. Then, when the specific area is selected through the terminal 300, the area-specific commercial value analysis module 172 counts the number of interest point morpresent corresponding to the selected area. If the weights are set to reflect the weights at the points of interest, the regional commercial analysis module 172 reflects the weights to the number of points of interest. In addition, the regional commercial analysis module 172 counts the number of location morpres- sories corresponding to the selected area. If the location morpheme is set to reflect the weight, the regional commercial analysis module 172 reflects the weight to the number of location morphemes. The regional commercial value analysis module 172 adds the number of the morpheme points of interest and the number of the position morphemes, and then compares the sum value and the reference value with each other. If it is checked that the total value is smaller than the reference value, the region-specific commercial analysis module 172 discriminates the selected region as a non-commercial commercial. If it is checked that the total value is larger than the reference value, the region-specific commercial value analysis module 172 determines the selected region as a popular commercial area. If it is checked that the sum value is the same as the reference value, the region-specific commercial area analysis module 172 discriminates the selected area as a congested commercial area.

The periodic merchandise analysis module 174 provides the terminal 300 with information for area selection through the interface module 179. [ Providing the information for the area selection to the terminal 300 may be omitted. For example, if the regional commercial analysis is programmed to be performed in real time or periodically, providing the information for the area selection to the terminal 300 may be omitted. Then, the periodic merchandise analysis module 174 provides the terminal 300 with information for selecting the first period and the second period, if the specific region is selected through the terminal 300. [ If it is checked that the first period and the second period are confirmed, the periodic commercial value analysis module 174 counts the number of morpheme corresponding to the selected region and the selected first period to generate a first count value. The periodic commercial analysis module 174 counts the number of morphemes corresponding to the selected region and the selected second period to generate a second count value. If the first count value is greater than the second count value, then the merchandise value analysis module 174 compares the first count value with the second count value. If the first count value is greater than the second count value, . If the first count value is less than the second count value, the periodical commercial value analysis module 174 determines that the selected region is a popular commercial area for the first period. If the first count value and the second count value are checked to be equal to each other, the periodical commercial value analysis module 174 determines that the selected region is a stagnant commercial region for the first period or the second period.

The merchandise analysis module 176 for each web document type provides information for selecting a web document to the terminal 300 through the interface module 179. [ Providing the information for selecting the web document to the terminal 300 may be omitted. For example, if it is programmed to perform commercial analysis by web document type in real time or periodically, providing the information for selecting the web document to the terminal 300 may be omitted. If it is checked through the interface module 179 that the specific web document is selected through the terminal 300, the merchandise value analysis module 176 for each web document type counts the number of points of interest corresponding to the selected web document . If the weights are set to reflect the weights at the interest point morpheme, the merchandise value analysis module 176 for each web document type reflects the weights to the number of morpheme points of interest. Then, the commercial document analysis module 176 counts the number of location morphemes corresponding to the selected web document. If the position morpheme is set to reflect the weight, the commodity analysis module 176 for each web document type reflects the weight to the number of position morphemes. The commercial document analysis module 176 for each web document type sums up the number of interest morphemes and the number of location morphemes to generate a sum value. The commercial document analysis module 176 for each web document type compares the generated sum value with predetermined reference values. If it is checked that the total value is smaller than the reference value, the merchandise analysis module 176 for each web document type determines that the area included in the selected web document is a non-merchant merchant. If it is checked that the sum is larger than the reference value, the merchandise analysis module 176 for each web document type determines that the region included in the selected web document is a popular merchandise. If the total value is checked to be equal to the reference value, the merchandise entry analysis module 176 for each web document type determines that the area included in the selected web document is a stadium area.

The trade point analysis module 178 for each point of interest category provides the terminal 300 with information for selecting a point of interest category via the interface module 179. Providing the terminal 300 with information for the above-mentioned interest point category selection may be omitted. For example, providing the terminal 300 with information for the above-mentioned interest point category selection may be omitted when the commercial analysis is performed to perform the commercial point-by-point category analysis in real time or periodically. When the particular interest point category is selected through the terminal 300 through the interface module 179, the commercial value analysis module 178 for each interest point category counts the number of interest point morpheme corresponding to the selected interest point category do. If the interest point morpheme is set to reflect the weight, the trade point analysis module for each point of interest 178 reflects the weight to the number of points of interest morpheme. The trade point analysis module 178 for each point of interest category counts the number of position morphemes corresponding to the selected point of interest category. If the location morpheme is set to reflect the weight, the trade point analysis module 178 for each point of interest category reflects the weight to the number of location morphemes. The merchandise analysis module 178 for each interest point category sums the number of interest morphemes and the number of position morphemes, and compares the summation value with a reference value. If it is checked that the sum value is smaller than the reference value, the commercial spot analysis module 178 of the interest spot category determines that the selected interest spot category is the non-commercial spot. If it is checked that the summation value is larger than the reference value, the commercial value analysis module 178 for each point of interest category determines that the selected point of interest category is the popular commercial value. If the summation value is checked to be equal to the reference value, the merchandise analysis module 178 for each interested point category determines that the selected interest point category is a congested merchant.

FIG. 3 is a flowchart illustrating a method of analyzing a local commercial floor flow according to an exemplary embodiment of the present invention.

Referring to FIG. 3, Web documents such as social commerce, performances, SNS, blog, news, and the like are collected and stored (step S100). The web document may be collected in real time or may be periodically collected at predetermined intervals.

Next, the morpheme is analyzed in the collected web document (step S102). In this embodiment, the morpheme may include a point morpheme corresponding to the information registered in the point of interest (POI) DB and a position morpheme corresponding to the position information not registered in the point of interest DB.

Then, one morpheme is extracted from the analyzed morphemes (step S104).

Then, it is checked whether or not the morpheme extracted in step S104 is the interest point morpheme (step S110).

If it is checked in step S110 that the point of interest is a morpheme of interest, the interest point morpheme is classified on a regional basis (step S112).

Then, the classified interest point morpheme is stored in the interest point morpheme DB (step S114).

On the other hand, if it is checked in step S110 that the morpheme is not the point of interest, it is checked whether or not the morpheme is a position morpheme (step S120).

If it is not checked that it is a position morpheme in step S120, the process returns to step S104. A morpheme that is neither a point of interest point nor a position morpheme can be a morpheme that refers to the name of a person or a morpheme that refers to a general object, such as a cup, a tray, a desk, and the like.

If it is checked in step S120 that the location morpheme is located, the location morpheme is classified on an area basis (step S122).

Then, the classified position morpheme is stored in the position morpheme storage DB (step S124).

It is checked whether there is a residual morpheme not extracted in the analyzed morphemes (step S130).

If it is checked in step S130 that the residual morpheme exists, the flow returns to step S104.

If it is checked in step S130 that the residual morpheme is not present, the commercial area is determined based on the number of morphemes, that is, the number of interest morpres and the number of position morpreses (step S140).

A large number of interest point morphemes or a large number of position morphemes means that many people generate and reproduce various web documents. Accordingly, it is possible to determine whether the commercial area of a specific area is a popular commercial area or an unlicensed commercial area. For example, in a particular area, if the number of interest point morphemes or the number of position morphemes is greater than the reference value, then the region can be identified as a popular commercial area, and if the number of interest morphemes or the number of position morphemes is smaller than the reference value, It can be determined as a commercial area.

Also, it is possible to grasp the flow of commerce in a specific area for a certain period of time. For example, if the number of interest point morphemes or the number of position morphemes corresponding to a specific region during the summer of this year corresponds to a specific region and is greater than the number of interest morphemes or position morphemes during the summer period of one year ago, It can be judged that it is floating in the area. On the other hand, if the number of interest branch morphemes or the number of location morphemes corresponding to a specific region during the summer of this year corresponds to a specific region and is smaller than the number of interest morphemes or position morphemes during the summer period of one year ago, It can be judged that it is being faded.

FIG. 4 is a flowchart for explaining the step of discriminating the commercial area shown in FIG.

Referring to FIG. 3 and FIG. 4, it is checked whether regional commercial analysis is requested (step S210).

If it is checked in step S210 that the regional merchandise analysis is requested, the regional merchandise is analyzed (step S220). The regional commercial analysis will be described in detail later with reference to FIG. In this embodiment, checking whether or not a request to analyze a commercial article by region can be made by a program configured to perform regional commercial analysis in real time or periodically. For example, it is determined that a request for analysis of a merchandise by region has been made according to the arrival of a specific time, and the above-described merchandise analysis for each region can be performed. Accordingly, it is possible to collect a large number of web documents in real time or at regular intervals, extract various interest point morphemes or location morphemes contained in the collected web documents, By using the information for analysis, it is possible to perform the commercial analysis operation according to the time, and the reliability of the commercial analysis can be improved.

If it is checked in step S210 that regional commercial analysis is not requested, it is checked whether or not a commercial analysis for each period is requested (step S230).

If it is checked in step S230 that the commercial article analysis by period is requested, the commercial article by period is analyzed (step S240). The above-mentioned period-by-period commercial analysis will be described in detail with reference to FIG. In this embodiment, checking whether or not the request for analysis of the commercial article according to the period is performed can be made by a program configured to perform regional commercial analysis in real time or periodically. For example, it is determined that the request for the analysis of the merchandise for each period is made according to the arrival of the specific time, so that the merchandise analysis operation for each period can be performed. Accordingly, it is possible to collect a large number of web documents in real time or at regular intervals, to extract various interest point morphemes or location morphemes contained in the collected web documents, By using the information for analysis, it is possible to perform the commercial analysis operation according to the time, and the reliability of the commercial analysis can be improved.

If it is checked in step S230 that the commercial article analysis by period is not requested, it is checked whether commercial article analysis by the type of web document is requested (step S250).

If it is checked in step S250 that the commercial document analysis by the type of the web document is requested, the commercial document by the web document type is analyzed (step S260). The above-described analysis of the commercial article according to the type of the web document will be described in detail with reference to FIG. In this embodiment, checking whether or not a commercial document analysis request for each type of web document is checked can be performed by a program configured to perform commercial document analysis for each type of web document in real time or periodically. For example, it is determined that a request for analyzing a piece of merchandise according to a type of a web document is made according to the arrival of a specific time, and the above-described merchandise analysis for each region can be performed. Accordingly, it collects a large number of web documents in real time or at regular intervals, extracts various interest point morphemes or location morphemes included in the collected web documents, and stores various information for commercial analysis according to the types of web documents based on the collected morphemes. By using the information for analyzing the commercial area, it is possible to perform the commercial analysis operation according to the time, and the reliability of the commercial area analysis can be improved.

If it is checked in step S250 that the commercial document analysis by web document type is not requested, it is checked whether or not the commercial document analysis by interest point category is requested (step S270).

If it is checked in step S270 that a commercial analysis by interest point category is requested, commercials by interest point category are analyzed (step S280). The commercial analysis by the interest point categories described above will be described in detail in FIG. 8 to be described later. In this embodiment, checking whether or not to request a commercial analysis by interest point category can be performed by a program set to perform commercial analysis by interest point category in real time or periodically. For example, as the specific time comes, it is determined that a request for analysis of a commercial article by category of interest has been made, so that the commercial analysis operation can be performed according to the region. Accordingly, it is possible to collect a large number of web documents in real time or at regular intervals, to extract various interest point morphemes or location morphemes contained in the collected web documents, Using the information for the accumulated commercial analysis, it is possible to perform the commercial analysis operation according to the time, and the reliability of the commercial analysis can be improved.

In the present embodiment, the commercial analysis is performed in the order of regional commercial analysis, commercial analysis by period, commercial analysis by web document type, and commercial analysis by interest branch category, but the order may be variously changed. In addition to the above-described commercial article analysis, various other types of commercial article analysis are also possible.

In addition, in the present embodiment, it has been explained that the steps of analyzing a commercial area by region, analyzing a commercial area by period, analyzing a commercial area according to a type of a web document, or analyzing a commercial area by category of interest are separately performed. However, The steps may be used mixed with each other. For example, it is possible to analyze regional commercial areas and periodical commercial areas, and analyze regional commercial areas, commercial areas by period, and commercial areas classified by types of web documents. It is also possible to analyze commercials by region, by period, by web document type, and by point of interest category.

FIG. 5 is a flowchart for explaining the regional merchandise analysis steps shown in FIG.

Referring to FIGS. 4 and 5, a specific region is selected (step S300).

Then, the number of points of interest morpheme corresponding to the selected area is counted (step S304).

Then, whether or not the weight is reflected in the morpheme of interest is checked (step S306). In this embodiment, the weights reflected in the morpheme of the interest point can be variously set by the manager in order to adjust the weight of the commercial analysis by the interest point morpheme and the weight of the commercial analysis by the position morpheme. For example, if we want to increase the weight of the commercial analysis by the interest branch morpheme rather than the commercial segment analysis by the location morpheme, the number of branch morphemes of interest is multiplied by the weight, for example, 1.1 or 1.5, To the number of location morphemes, it is possible to increase the proportion of commercial analysis by the morpheme of interest. In order to reduce the proportion of commercial analysis by the interest branch morpheme, the number of branch morphemes of interest is multiplied by a weighting factor, for example, 0.8 or 0.9, The number of morphemes can be lowered, and the proportion of commercial analysis by location morphemes can be increased.

If weights are checked in step S306, the weight is reflected in the number of points of interest morpheme (step S308).

In step S306, the number of position morpres- sories corresponding to the selected area is counted (step S310), or after the step S308, the number of position morpres- sories corresponding to the selected area is counted.

Whether or not the position morpheme reflects the weight is checked (step S312). In this embodiment, the weights reflected in the location morpheme can be variously set by the administrator to adjust the weight of the commercial analysis by the interest point morpheme and the weight of the commercial analysis by the location morpheme. For example, if we want to increase the proportion of commercial analysis by location morpheme rather than the proportion of commercial analysis by the interest branch morpheme, we can estimate the number of position morphemes by multiplying the position morpheme with weights, for example, 1.1 or 1.5, The number of branch morphemes is higher than that of branch morphemes. In order to reduce the proportion of commercial analysis by location morpheme rather than the proportion of commercial analysis by the interest branch morpheme, the number of position morphemes is multiplied by the weight, eg, 0.8 or 0.9, It is possible to increase the proportion of commercial analysis by the morpheme of interest.

In step S312, the weight is reflected in the number of position morphemes (step S314).

In step S312, the number of interest morpres and the number of position morpreses are added (step S316).

The sum value S and the reference value R are compared with each other (step S318).

If it is checked in step S318 that the sum value S is smaller than the reference value R, the selected area is determined as an unauthorized commercial area (step S320).

If it is checked in step S318 that the sum value S is larger than the reference value R, the selected area is determined as a popular commercial area (step S322).

If it is checked in step S318 that the sum value S is the same as the reference value R, the selected area is determined to be a still image area (step S324).

FIG. 6 is a flow chart for explaining the commercial article analysis step according to the period shown in FIG.

Referring to FIGS. 4 and 6, a specific region is selected (step S400).

Next, the first period and the second period are respectively selected (step S404). In this embodiment, the first period may be set to be earlier than the second period. For example, if the second period is September 2014, the first period may be set to September 2013. Although it has been described in the present embodiment that information for period selection is provided after information for area selection is provided, two types of information may be provided together through one user interface screen.

Then, it is checked whether the first period and the second period are confirmed for the commercial analysis through the terminal (step S406).

If it is checked in step S406 that the first period and the second period are determined, the number of morpresent corresponding to the selected region and the selected first period is counted to generate the first count value C1 (step S408).

Next, the number of morphemes corresponding to the selected region and the selected second period is counted, and a second count value C2 is generated (Step S410).

The first count value C1 and the second count value C2 are compared with each other (step S412).

If it is checked in step S412 that the first count value C1 is larger than the second count value C2, the selected area is determined as a popular commercial area for the second period (step S414). For example, if the selected region is Gangnam Station, the first period is September 2013, and the second period is September 2014, the area near Gangnam Station will rise to popular trade in September 2014 compared to September 2013 .

If it is checked in step S412 that the first count value C1 is smaller than the second count value C2, the selected area is determined as a popular commercial area for the first period (step S416). For example, if the selected region is Gangnam Station, the first period is September 2013, and the second period is September 2014, the area near Gangnam Station will be lowered to the non-commercial area in September 2014 compared to September 2013 .

If it is checked in step S412 that the first count value C1 and the second count value C2 are the same, the selected area is determined to be a congested area for the first period or the second period (step S418). For example, if the selected region is Gangnam Station, the first period is September 2013, and the second period is September 2014, the area near Gangnam Station will be September 2013 or September 2014, It can be judged that it is maintaining the congestion state.

FIG. 7 is a flow chart for explaining the merchandise analysis step for each web document type shown in FIG.

Referring to FIGS. 4 and 7, a specific web document is selected (step S500). For example, any one of web documents such as social commerce, performance, SNS, blog, and news may be selected. Here, a specific area may be selected additionally.

Then, the number of points of interest morpheme corresponding to the selected web document is counted (step S504).

Then, whether or not the weight is reflected in the morpheme of interest is checked (step S506). In this embodiment, the weights reflected in the point of interest morpheme are described in Fig. 5, and a description thereof will be omitted.

If weights are checked in step S506, weights are reflected in the number of points of interest morphology (step S508).

In step S506, the number of position morphemes corresponding to the selected web document is counted (step S510).

Whether or not the position morpheme reflects the weight is checked (step S512). In this embodiment, the weight values reflected in the position morpheme are described in Fig. 5, and a description thereof will be omitted.

In step S512, the weight is reflected in the number of position morpreses (step S514).

In step S512, the number of points of interest and the number of position morpreses are summed to generate a summation value S (step S516).

The sum value S generated in step S516 is compared with a preset reference value R (step S518).

If it is checked in step S518 that the sum value S is smaller than the reference value R, the selected web document is determined as an unauthorized merchandise (step S520).

If it is checked in step S518 that the sum value S is larger than the reference value R, the selected web document is determined to be a popular commercial article (step S522).

If it is checked in step S518 that the sum value S is the same as the reference value R, the selected web document is judged to be a still image area (step S524).

FIG. 8 is a flowchart for explaining a commercial area analysis step for each point of interest category shown in FIG.

Referring to Figs. 4 and 8, a specific point of interest category is selected (step S600). In the present embodiment, the interest point category can be defined by being divided into a large category, a middle category classified in one large category, and a small category classified in one middle category. For example, the major category can be divided into life, shopping, and the like. The middle category can be divided into restaurants and large retail stores. The subcategories can be divided into fast food, department stores, etc.

Then, the number of points of interest morpresent corresponding to the selected interest point category is counted (step S604).

Then, whether or not the weight is reflected in the morpheme of interest is checked (step S606). In this embodiment, the weights reflected in the point of interest morpheme are described in Fig. 5, and a description thereof will be omitted. If weights are checked in step S606, the weight is reflected in the number of points of interest morpheme (step S608).

In step S606, the number of position morphemes corresponding to the selected interest point category is counted (step S610).

Whether or not the position morpheme reflects the weight is checked (step S612). In this embodiment, the weight values reflected in the position morpheme are described in Fig. 5, and a description thereof will be omitted. In step S612, the weight is reflected in the number of position morpreses (step S614).

In step S612, the number of interest morpres and the number of position morpreses are added (step S616).

The sum value S and the reference value R are compared with each other (step S618).

If it is checked in step S618 that the sum value S is smaller than the reference value R, the selected interest point category is determined as an unauthorized commercial area (step S620).

If it is checked in step S618 that the sum value S is larger than the reference value R, the selected interest point category is determined as a popular commercial area (step S622).

If it is checked in step S618 that the sum value S is the same as the reference value R, the selected interest point category is determined to be a congested area (step S624).

Hereinafter, the commercial analysis according to the present embodiment will be described in more detail. For convenience of explanation, we set Gangnam-gu, Sinsa-dong, Yeongdeungpo, and Myung-dong, which are important commercial areas in Seoul, as the region, and the period of extracting the interest point morpheme and location morpheme from the web document is from January 2014 to November 2014 Extraction will be described as an example.

Table 1 is a table for describing the total number of interest point morphemes and position morphemes extracted by region and region in web documents. 9 is a graph for explaining the total number of interest point morphemes and location morphemes extracted by region and region in a web document. In particular, the graph corresponding to Table 1 is shown. In the present embodiment, an example of the total number of local morpheme (POI) morphemes and location morphemes extracted from a web document is shown.

[Table 1]

Referring to Table 1 and FIG. 9, in January 2014, the total number of points of interest and location morphemes related to Sinsa-dong, Yeongdeungpo and Myung-dong in the web document were 1,000, 3,000 and 5,000, The total number of morpheme and location morphemes related to Sinsa-dong, Yeongdeungpo, and Myung-dong in the document are 1,000, 3,100, and 4,500, respectively. In March 2014, each of the points of interest related to Sinsa-dong, Yeongdeungpo, And the total number of location morphemes is 1,000, 3,000 and 5,000.

Also, in April of 2014, the total number of interest point morphemes and location morphemes related to Sinsa-dong, Yeongdeungpo and Myung-dong in Web documents were 1,000, 2,900 and 5,200, respectively. In May 2014, The total number of morphemes and location morphemes for each interest point related to Sinsa-dong, Yeongdeungpo, and Myung-dong in the web document in June 2014 is 1,000, 2,800 and 5,200, 1,000, 2,700, 5,100.

In this way, the total number of interest point morphemes and location morphemes related to each region in the web document can be arranged corresponding to each July to November, 2014.

Referring to FIG. 9, a web document analysis index can be defined based on the total number of interest morphemes and location morphemes extracted corresponding to each region in each month in web documents.

As shown in FIG. 9, even though the time elapses, it can be confirmed that Myungdong is hit the most in a web document, Yeongdeungpo is hit at a medium level, and the sinsomong is hit to the lowest level.

Table 2 is a table for explaining an example of the total number of points of interest (POI) morphemes corresponding to the respective regions in Table 1. 10 is a graph for explaining an example of the total number of points of interest (POI) morphemes corresponding to each area. In particular, the graph corresponding to Table 2 is shown.

[Table 2]

Referring to Table 2 and FIG. 10, in January 2014, the total number of interest point spots related to Sinsa-dong, Yeongdeungpo, and Myong-dong in web documents was 500, 400 and 3,500, and in February 2014, , Yeongdeungpo, and Myungdong are 480, 500, and 3,200, respectively. In March 2014, the total number of morpheme points of interest related to Sinsa-dong, Yeongdeungpo, 600 and 3,500, respectively.

Also, in April of 2014, the total number of morpheme points of interest related to Sinsa-dong, Yeongdeungpo, and Myeong-dong were 479, 650 and 3,200, respectively. In May 2014, web documents related to Sinsa-dong, Yeongdeungpo, The total number of each interest point morpheme is 500, 850, and 3,500, and in June 2014, the total number of interest point morphemes related to Sinsa-dong, Yeongdeungpo, and Myung-dong in web documents is 530, 900, and 3,600.

In this way, the total number of interest point morphemes related to each region in the web document can be arranged corresponding to each July to November, 2014.

As shown in FIG. 10, it is possible to define a local POI index based on the total number of interest point morphemes extracted corresponding to each region of each month in web documents. As shown in FIG. 10, it can be seen that even though the time elapses, Myongdong is the most hits in the web document and the sinsomong is the least hit. On the other hand, as time elapses, Yeongdeungpo is gradually getting hit.

On the other hand, in analysis of commercial areas, not only the popularity and the total number of shops but also the area and density of commercial areas are also important. In other words, since commercial districts are located not only on the first floor but also on two or more floors or in the basement, the area and density of the commercial area are variables to be considered when analyzing the commercial area.

In particular, stereoscopic analysis can enhance the reliability of commercial analysis rather than planar analysis. In order to analyze a stereoscopic commercial article, it is required to convert a two-dimensional (hereinafter, 2D) area into a three-dimensional (3D) area.

Table 3 shows examples of 2D area and 3D area in each area. Here, the area is simply represented by numbers so that relative size comparison is possible. Also, the 3D area can be calculated using the building information DB in the area. In Table 3, AOI means the area of interest.

[Table 3]

As shown in Table 3, it can be seen that Sinsa-dong is the smallest in the 2D area, Yeongdeungpo is the middle, and Myeongdong is the widest area, while the 3D area has the widest sinsa-dong, Myongdong is the middle and Yeongdeungpo is the smallest area. In other words, although the area of 2D is larger than that of Yeongdeungpo, Myungdong. However, there are relatively low - height buildings in Myung - dong, while relatively high - height buildings exist in Yeongdeungpo. Therefore, it can be seen that the difference between Yeongdeungpo and Myongdong is narrowed in the 3D area rather than the difference between Yeongdeungpo and Myongdong in 2D area.

In Table 3, the 2D area or the 3D area for each region may be fixed or changed over time. For example, if the current building is stretched or expanded relative to the previous one, the 2D area or 3D area may increase over time. However, in the present embodiment, for convenience of explanation, the 2D area or the 3D area of time is exemplified as being fixed.

Table 4 shows examples of densities of points of interest (POIs) per 2D area area in each area. 11 is a graph for explaining an example of the density of the point of interest (POI) per 2D unit area of each area. In particular, the graph corresponding to Table 4 is shown. In this embodiment, the total number of interest point morpheme points for each region shown in Table 2 divided by the 2D area for each region shown in Table 3 is defined as the density of the point of interest per 2D unit area.

[Table 4]

Referring to Table 4 and FIG. 11, the density of the points of interest per each 2D unit area related to Sinsa-dong, Yeongdeungpo, and Myondong during Jan 2014 is 0.5, 0.1333, and 0.7, The density of the points of interest per each 2D unit area is 0.48, 0.1666, and 0.64, and the density of the points of interest per each 2D unit area related to Sinsa-dong, Yeongdeungpo, and Myondong during March 2014 is 0.52, 0.2, and 0.7.

In this way, the density of points of interest per 2D unit area of each region can be calculated corresponding to April to November 2014 respectively.

11, there is no change in the density of the point of interest per 2D unit area corresponding to the sphere of Myeongdong or the density of the point of interest per 2D unit area corresponding to the sphinx of the soul. However, Is gradually increased.

Since the density of the point of interest per 2D unit area means that the density of the commercial area is low, it can be deduced that the commercial density is low as the density of the commercial area is low. On the other hand, high density of points of interest per 2D unit area means high densities of shopping malls, so it can be inferred that the higher the density of downtown malls, the more fierce the competition is.

Table 5 shows an example of the density of the point of interest (POI) per 3D area of the area. FIG. 12 is a graph for explaining an example of the density of the point of interest (POI) per 3D unit area of the area. In particular, a graph corresponding to Table 5 is shown. In this embodiment, the total number of morpheme points of interest for each region shown in Table 2 divided by the 3D area for each region shown in Table 3 is defined as the density of the point of interest per 3D unit area.

[Table 5]

Referring to Table 5 and FIG. 12, the densities of the points of interest per each 3D unit area related to Sinsa-dong, Yeongdeungpo, and Myondong during January 2014 are 0.05, 0.08 and 0.5833, and in Sinsa-dong, Yeongdeungpo, The densities of the points of interest per each 3D unit area are 0.048, 0.1 and 0.5333, and the densities of the points of interest per each 3D unit area related to Sinsa-dong, Yeongdeungpo, and Myong-dong in March 2014 are 0.052, 0.12 and 0.5833.

In this way, the density of points of interest per 3D area of each area can be calculated corresponding to April to November 2014 respectively.

As can be seen in FIG. 12, it can be seen that the density of the point of interest per 3D unit area corresponding to the sphere of Myeongdong is always higher than the density of the point of interest per 3D unit area corresponding to the sphincter. It can also be seen that there is no significant change in the density of the point of interest per 3D unit area corresponding to the myeongdong, or the density of the point of interest per 3D unit area corresponding to the sphincter, with the passage of time.

However, it can be seen that as the time passes, the density of the point of interest per 3D unit area corresponding to Yeongdeungpo gradually increases. In particular, during January 2014, the density of points of interest per unit area of 3D was almost the same for Sinsa-dong and Yeongdeungpo, but over time, the density of points of interest per 3D unit area increased during November 2014, .

The low density of the point of interest per 3D unit area means that the density of the commercial area is low, so it can be deduced that the commercial density is low as the density of the commercial area is low. On the other hand, high density of points of interest per 3D unit area means high densities of shopping malls, so it can be inferred that the higher the densities of shopping malls, the more fierce the competition is.

Based on the above-described tables, the simple commercial index, the 2D commercial index reflecting the 2D area, and the 3D commercial index reflecting the 3D area can be calculated as shown in Tables 6, 7 and 8, respectively.

Table 6 is a table for explaining simple commercial indexes. 13 is a graph for explaining the simple commercial exponent index. In particular, a graph corresponding to Table 6 is shown. In the present embodiment, a value obtained by dividing the total number of points of interest morpheme and position morpheme by period for each region shown in Table 1 is divided by the total number of POI morphemes shown in Table 2 as a commercial index.

[Table 6]

Referring to Table 6 and FIG. 13, the simple commodity indexes corresponding to Sinsa-dong, Yeongdeungpo and Myondong in January 2014 are 2, 7.5 and 1.428, respectively. In February 2014, simple commercial indexes corresponding to Sinsa-dong, Yeongdeungpo, While the simple commercial index corresponding to Sinsa-dong, Yeongdeungpo, and Myeong-dong is 1.923, 5, and 1.428, respectively, in March 2014.

In this way, simple trade indexes for each region can be calculated corresponding to April to November 2014 respectively.

As can be seen in FIG. 13, it can be seen that the simple commercial index of Myondong is relatively low overall, and there is no change. However, we can confirm that the simple commercial index of Sinsa-dong gradually increases with time. In addition, Yeongdeungpo's simple commercial index is very high in January 2014, but it shows that it drops sharply over time.

Table 7 is a table for explaining the 2D commercial index reflecting the 2D area. FIG. 14 is a graph for explaining the 2D commercial index reflecting the 2D area. In particular, a graph corresponding to Table 7 is shown. In this embodiment, the total number of interest point morpheme and position morpheme for each region shown in Table 1 is divided by the total number of POI morphemes shown in Table 2, The value divided by the density of the point of interest is defined as a trade index.

[Table 7]

Referring to Table 7 and FIG. 14, the 2D commercial index corresponding to Sinsa-dong, Yeongdeungpo, and Myeong-dong is 4, 56.25, and 2.0408 in January 2014, and the 2D commercial index corresponding to Sinsa-dong, Yeongdeungpo, Respectively, and the 2D commercial index corresponding to Sinsa-dong, Yeongdeungpo, and Myeong-dong is 3.698, 25, and 2.040 in March 2014, respectively.

In this manner, 2D commercial indexes for each region can be calculated corresponding to April to November 2014 respectively.

As can be seen from FIG. 14, the overall 2D commercial index of Myungdong is relatively low, and it can be confirmed that there is no change. However, it can be seen that the 2D commercial index of Sinsa-dong gradually increases with time. In addition, the 2D commercial index of Yeongdeungpo is very high in January 2014, but it shows that it drops sharply over time.

That is, the 2D commercial index increases as the 2D density decreases, and the 2D commercial index decreases as the 2D density increases. A low 2D density may mean that the number of competitors is small, which means that the area covered by one shop is wide.

Table 8 is a table for explaining 3D commercial index reflecting 3D area. FIG. 15 is a graph for explaining the 3D premium index reflecting the 3D area. In particular, the graph corresponding to Table 8 is shown. In the present embodiment, the total number of interest point morpheme and position morpheme per each region shown in Table 1 is divided by the total number of POI morphemes shown in Table 2, The value divided by the density of the point of interest is defined as a trade index.

[Table 8]

Referring to Table 8 and FIG. 15, 3D commercial indexes corresponding to Sinsa-dong, Yeongdeungpo and Myondong are 40, 93.75 and 2.4489, respectively, during January 2014, and 3D commercial indexes corresponding to Sinsa-dong, Yeongdeungpo, Respectively, and the 3D commercial index corresponding to Sinsa-dong, Yeongdeungpo, and Myeong-dong is 36.98, 41.66, and 2.4489 in March 2014, respectively.

In this way, 3D commercial indexes for each region can be calculated corresponding to April to November 2014 respectively.

As can be seen in FIG. 15, the overall 3D commercial index of Myondong is very low compared to the other two regions, and it can be seen that there is no change. On the other hand, Yeongdeungpo 3D commercial index is very high in January 2014, but it shows a rapid decline over time. Nevertheless, the 3D commercial index of Yeongdeungpo is higher than the 3D commercial index of Myongdong.

However, the 3D commercial index of Sinsa-dong is higher than the 3D commercial index of Myong-dong and the 3D commercial index of Yeongdeungpo from January 2014 to March 2014, but it is higher than the 3D commercial index of Myong-dong and Yeongdeungpo from April 2014 . In other words, the 3D commercial index of Sinsa-dong gradually increases with time, which is much higher than the 3D commercial index of Myeong-dong and Yeongdeungpo in November 2014.

That is, the 3D commercial index increases as the 3D density decreases, and the 3D commercial index decreases as the 3D density increases. A low 3D density may mean that the number of competitors is small, which means that the area covered by a single mall is wide.

The following are briefly summarized as follows.

Referring to FIG. 9, if there is a lot of mention in a web document, the area appears to be a popular commercial area. However, referring to FIG. 13, the relative value of the commercial area can be known by the web analysis on the time series and the POI density. In other words, even if the results of web analysis are good, Yeongdeungpo and Myungdong can be seen to be rather bad when the POI is considered.

Also, referring to FIG. 14, it is possible to analyze from another viewpoint considering the area density of the area. In other words, although the commercial area seems to be getting better in Fig. 13, considering the density, it can be confirmed that the commercial area does not improve very much especially in the case of Sinsa-dong.

Also, considering the 3D density considering the local building layer, it is possible to analyze another viewpoint again. That is, in FIG. 14, it can be confirmed that the sinsa-dong commercial area which does not look very good is remarkably improved by checking Fig. This is due to the high density of the high-rise buildings and the relatively wide availability of the area.

The embodiments of the present invention described above can be implemented in the form of program instructions that can be executed through various computer components and recorded in a computer-readable recording medium. The computer-readable recording medium may include program commands, data files, data structures, and the like, alone or in combination. The program instructions recorded on the computer-readable recording medium may be those specifically designed and configured for the present invention or may be those known and used by those skilled in the computer software arts. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape, optical recording media such as CD-ROM and DVD, magneto-optical media such as floptical disks, medium, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code, such as those generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware device may be modified into one or more software components for performing the process according to the present invention, and vice versa.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention as defined in the appended claims. You will understand.

As described above, according to the present invention, attention is paid to the fact that regional information and regional characteristics are analyzed as a flow of time, and commercial analysis is required for the region. Thus, real-time web information such as social commerce, performance, SNS, It collects documents and extracts location and point of interest (POI) from the collected web documents, classifies and analyzes them on a regional basis, and determines popular areas based on the number of times mentioned. In particular, it is possible to determine which category is popular depending on the type of the extracted web document and the POI category.

100: Local commerce flow analysis system 110: Web document collection section
120: web document storage DB 130: morphological analysis unit
140: morpheme classification-storing section 142: morpheme extracting section
144: morpheme classifier 150: store point of interest morpheme DB
160: location morpheme storage DB 170: merchandise flow discrimination unit
172: Regional trade analysis module 174: Regional trade analysis module
176: Commercial analysis module by web document type
178: Commercial analysis module by category of interest
179: Interface module 200: POI DB
300:

Claims (12)

Collecting a plurality of web documents;
Performing morpheme analysis on the collected web documents;
Classifying and analyzing the analyzed morphemes by region; And
And determining a commercial area based on the number of morphemes stored on the area basis. 2. The system according to claim 1, wherein the morpheme includes a point morpheme corresponding to the information registered in the point of interest (POI) DB and a position morpheme corresponding to the position information not registered in the point of interest DB How to analyze commercial flow. 3. The method of claim 2, wherein classifying and storing the analyzed morphemes by region comprises:
Extracting the interest point morpheme or the position morpheme in the analyzed morphemes; And
And classifying and storing each of the interest point morpheme and the position morpheme on a regional basis. 4. The method according to claim 3, wherein the step of discriminating the commercial area comprises: determining the commercial area based on the number of the interest point morpres and the number of the location morpreses stored on the area basis . 5. The method of claim 4, wherein the step of determining the commercial area comprises:
Analyzing the commercial area according to the region, analyzing the commercial area according to the region, analyzing the commercial area according to the region and the period, analyzing the commercial area according to the type of the web document written in correspondence with the region, And analyzing the commercial according to the branch category. The method of claim 1, wherein the web document includes a social commerce, a performance, an SNS, a blog, and a news. A web document collection unit for collecting a plurality of web documents;
A morphological analysis unit for analyzing morphemes in the collected web documents;
A morpheme classifying and storing unit for classifying and analyzing the analyzed morphemes by region; And
And a merchandise flow discrimination unit for discriminating a merchandise area based on the number of morphemes stored on the basis of the area standard. 8. The method according to claim 7, wherein the morpheme includes a point morpheme corresponding to the information registered in the point of interest (POI) DB and a position morpheme corresponding to the position information not registered in the point of interest DB Commercial flow analysis system. 9. The apparatus according to claim 8, wherein the morpheme classification-
A morpheme extracting unit for extracting the morpheme of interest and the morpheme corresponding to the position information in the analyzed morphemes; And
And a morpheme classifying unit for classifying each of the interest point morpheme and the position morpheme on an area basis. 10. The method of claim 9,
A point of interest morpheme storage DB storing an interest point morpheme classified by the morphological classification unit; And
And a location morpheme storage DB for storing location morphemes classified by the morpheme classification unit. 11. The apparatus according to claim 10,
Characterized in that the commercial area is determined by analyzing the flow of time on the basis of the number of interest point morphemes stored in the interest point morpheme storage DB and the number of location morpreses stored in the location morpheme store DB Commercial flow analysis system. 12. The apparatus according to claim 11,
A commercial analysis module for analyzing the commercial area according to the region, a commercial analysis module for analyzing the commercial area in correspondence with the region and the period, a commercial analysis module for the web document classified by the type of the web document written corresponding to the region, And a commercial analysis module for each point of interest that analyzes commercials by branch category.

Images