JP2011221693A - Method, apparatus and program for detecting change in geographical attribute - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect a geographical attribute whose update is necessary and to keep the geographical attribute in a latest state at low cost.SOLUTION: A change detection part 11 detects a change in a value of a geographical attribute item, a related change detection part 12 retrieves relation between geographical attribute items including the geographical item whose change is detected from an item relation storage part 13 to detect a geographical attribute item whose change is predicted. Consequently, a geographical attribute item belonging to a difficult update group whose inspection is necessary can be grasped and the geographical attribute can be kept in the latest state.

Description

  The present invention relates to a technology for efficiently updating peripheral geographical features related to an outdoor place such as a store, a station, or a city.

  In recent years, ICT services for outdoor spaces such as advertisement distribution solutions for digital signage and search sites that introduce nearby stores using location information have been expanded. In the outdoor space ICT service, if the tendency of attributes of people (hereinafter referred to as “users”) gathered in the place as well as location information, a list of nearby stores, etc. can be grasped, information more suitable for the user is presented. Can be effective. For example, if you have a lot of users who are going home after work, you may want to present an advertisement for alcoholic beverages that will increase consumption on the way home from work, or stop by with friends if you have a lot of users coming to play with friends. By presenting advertisements for many cafes, snacks, and restaurants, it is possible to present information more suited to the user's situation.

  In order to grasp the tendency of the attribute of the user, for example, as in Non-Patent Document 1, a surveyor is dispatched to a place where the user wants to know the tendency, and a questionnaire or interview survey is performed on the user. There is a method of acquiring data by conducting a web or mail survey for a user who has visited the site. However, in order to grasp the tendency of the attribute of the user in many places of tens to several hundreds or more, which are the targets of the ICT service for outdoor space, the method of performing the above survey takes a lot of cost and labor. There is a problem.

  In order to grasp the user's attributes, the geographical features around the location related to the user's attributes can be used. In Non-Patent Document 1, the difference in the user attribute due to the difference in the location was verified by targeting the theater in the city area and the main street as the location, and the characteristics of the buying behavior of the visitor as the user attribute. In the verification of Non-Patent Document 1, it was shown that the difference in the location brings about the difference in the buying behavior pattern. The study of Non-Patent Document 1 was a case study, but discovered and used general relationships between geographical features (area characteristics such as theaters and main streets) and user attributes (buying behavior patterns). By doing so, it is considered that the attribute of the user with high acquisition cost can be estimated by the geographical feature that can be easily acquired.

  Hereinafter, user attributes are referred to as “user attributes”, and geographical features are referred to as “geographic attributes”.

  Give examples of user attributes and geographic attributes, and the relationship between them. For example, in a business district (geographic attribute) where many offices gather, it is assumed that there are many business people (user attributes) who return home after work from evening to night. In another example, it is assumed that there are many people who come to play with friends (user attributes) in places (geographic attributes) where there are many restaurants and pubs. Thus, by estimating the user attribute from the geographic attribute, it is possible to minimize the cost of directly investigating the user attribute.

  FIG. 9 shows an example of the user attribute and geographic attribute of a certain place.

  The “place” refers to a rough range such as Otemachi, Marunouchi, Yurakucho, etc., for example, within a radius of several hundred meters centered on the station and imagined when a person hears the place name.

  The “user attribute” is information about the user's behavior and attribute for the purpose of helping to send information to the place. In addition to gender, age, and occupation, assume trends in purpose of visit and mood.

  "Geographical attributes" describe the location with objective indicators, such as the number of commercial facilities such as restaurants, the number of real estate properties, the number of higher education facilities such as universities, the number of entertainment facilities such as movie theaters, or It refers to the number of peripheral facilities such as the number of companies and houses, and static and steady population and economic indicators such as daytime population, staff per household, and number of households.

  Among the geographical attributes shown in FIG. 9, the number of restaurants, the number of real estate properties, the number of hotels, etc. can be easily referred to a frequently updated store database that can be used for services on the Internet. You can know the number. For example, a typical house map providing service including store information is regularly updated once every three or four months.

  On the other hand, static and steady population / economic indicators such as daytime population, staff per household, etc. are published by, for example, “National Census” published by the Ministry of Internal Affairs and Communications, “Office / Corporate Statistics”, and Ministry of Economy, Trade and Industry. It can be obtained from “Commercial Statistics” and “Industrial Statistics”. Since these surveys are conducted only once every five years, including simple surveys, it may take up to five years to update geographic attributes.

JP 2009-015796 A

Akihiko Misaka and Toshiyuki Kaneda, “Survey and analysis of the characteristics of the migratory behavior of theater visitors in central Tokyo”, Architectural Institute of Japan Technical Report, June 2005, No. 21, p.303-307 Genichiro Kikui, 4 others, “Rich annotation: Trial of knowledge extraction focusing on specific expressions”, IEICE Technical Report, NLC, July 2008, 108, 141, p.73 -78

  However, although population and economic indicators are static and steady, they vary from several months to several years, but change over time. Changes in geographic attributes often have a significant impact on user attributes. For example, if a location is redeveloped and a cinema complex with a large number of tenants starts operating, the number of people coming to play on weekends will change user attributes, and a tower apartment where many households can live If the building is constructed, the number of residents increases and user attributes such as the age of nearby places may change.

  As described above, the geographic attribute includes items having various update frequencies over a span of several months to five years. Geographic attributes do not change frequently in a short period of time, but once they change, the user attributes will be greatly affected, so updating the geographic attributes at the right time is important for grasping the appropriate user attributes. That is. Although it is possible to acquire the geographical attribute by hand, there is a problem that it becomes a very laborious and costly work especially when the geographical attribute is diverse.

  The present invention has been made in view of the above, and an object of the present invention is to detect a geographic attribute that needs to be updated and keep the geographic attribute up-to-date at a low cost.

  The geographic attribute change detection method according to the first aspect of the present invention refers to a database storing a first geographic attribute item indicating a predetermined attribute related to a place by a change detection means, and the value of the first geographic attribute item is determined. There is a correlation between the first geographic attribute item and the first geographic attribute item detected by the step of detecting a change and the related change detection means, and the attribute of the first geographic attribute item is different. The second geographic attribute item related to the first geographic attribute item detected in the step of detecting the change with reference to the relationship storage means that stores the relationship with the second geographic attribute item indicating the attribute. And a step of searching.

  In the geographic attribute change detection method, a step of inputting a text article by an article input means, and a place name in which place names for detecting changes in values of the first and second geographic attribute items by an article selection means are accumulated. A text article including a place name in which a change is detected in the first geographic attribute item in the step of selecting a text article including the place name stored in the storage means and the step of detecting the change by a keyword extracting means. Extracting the keyword representing the topic of the text article from the step, and extracting the second geographic attribute item and the keyword retrieved in the step of retrieving the second geographic attribute item by the updating means And storing the keyword in the updated geographic information storage means in association with the keyword. To.

  In the geographic attribute change detection method, the history of the first and second geographic attribute items and their values from the past geographic attribute storage means for storing the history of the first and second geographic attribute items and their values by the relationship extracting means. And calculating the correlation between the first geographic attribute item and the second geographic attribute item, and the relationship extraction means for the first geographic attribute item and the first geographic attribute item. And extracting the relationship with the correlated second geographic attribute item and storing it in the relationship storage means.

  In the geographic attribute change detection method, the related change detection unit refers to the relation storage unit, and in the step of detecting the change, another change associated with the place where the change of the first geographic attribute item is detected. And searching for the second geographic attribute item related to the first geographic attribute item in a place.

  The geographic attribute change detecting device according to the second aspect of the present invention has a correlation between a first geographic attribute item indicating a predetermined attribute relating to a location and the first geographic attribute item, and the first geographic attribute item. Reference is made to the relationship storage means for storing the second geographic attribute item indicating an attribute different from the attribute of the attribute item and the database storing the first geographic attribute item, and the first geographic attribute item is referred to A change detecting means for detecting a change in the value of the value, and a related change for searching for the second geographic attribute item related to the first geographic attribute item detected by the change detecting means with reference to the relation storage means And detecting means.

  In the geographical attribute change detecting device, a place name accumulating means for accumulating a place name for detecting a change in the value of the first and second geographic attribute items, an article input means for inputting a text article, and the place name accumulating An article selecting means for selecting a text article including a place name stored in the means, and a text article including the place name in which the change is detected in the first geographic attribute item in the step of detecting the change. A keyword extracting means for extracting a keyword representing a topic, the second geographic attribute item searched by the related change detecting means, and the keyword extracted by the keyword extracting means are associated with each other and stored in the update geographic information accumulation means And updating means.

  In the geographic attribute change detecting device, the first and second geographic attribute items and the past geographic attribute accumulation means for accumulating the history of the values, the first and second geographic attribute items from the past geographic attribute accumulation means, and the A history of values is read, the correlation between the first geographic attribute item and the second geographic attribute item is obtained, and the first geographic attribute item and the first geographic attribute item are correlated. And a relationship extracting unit that extracts a relationship with the second geographic attribute item and stores the relationship in the relationship accumulating unit.

  In the geographical attribute change detection device, the related change detection means refers to the relation storage means, and the first change is made in another place associated with the place where the change of the first geographical attribute item is detected. The second geographic attribute item related to the geographic attribute item is searched.

  A geographic attribute change detection program according to a third aspect of the present invention causes a computer to execute the geographic attribute change detection method.

  According to the present invention, a geographic attribute that needs to be updated can be detected, and the geographic attribute can be kept up to date at a low cost.

1 is an overall configuration diagram of a system including a geographic attribute change detection device in the present embodiment. It is a functional block diagram which shows the structure of the geographic attribute change detection apparatus in this Embodiment. It is a figure which shows the example of the relationship between geographic attribute items. It is a figure which shows the example of the place which a place name storage part accumulate | stores. It is a figure which shows the example of the update required geographic attribute information which the update required geographic attribute storage part accumulated. It is a flowchart which shows the flow of the process which detects the geographical attribute item which needs an update. It is a flowchart which shows the flow of the process which extracts a keyword. It is a flowchart which shows the flow of the process which extracts the relationship between geographic attribute items. It is a figure which shows the example of the user attribute and geographical attribute of a certain place.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is an overall configuration diagram of a system including a geographic attribute change detection device according to the present embodiment. The geographic attribute change detection device 1 can be connected to the network 100 and the user attribute estimation device 200. The user attribute estimation device 200 is a device that estimates a user attribute using the geographic attribute accumulated in the geographic attribute change detection device 1.

  The geographic attribute change detection device 1 acquires the value of the geographic attribute item from the store database 301, the real estate property database 302, etc. arranged on the network 100, detects the change, and uses the relationship between the geographic attribute items, Find another geographic attribute item to update.

  The geographic attribute item is information indicating a predetermined attribute related to a place such as the number of restaurants, the number of offices, the daytime population, and the land price, and the geographic attribute item has a value in each place. Although the actual value of the geographic attribute item changes with time, the value of the geographic attribute item accumulated in the geographic attribute change detection device 1 is acquired and updated from the store database 301, the real estate property database 302, or the like. It will be updated by the survey. Geographic attribute items with a high update frequency that can be easily acquired are called easy update groups, and geographic attribute items with a low update frequency are called hard update groups. In the example of the geographic attribute shown in FIG. 9, each geographic attribute item of the number of surrounding facilities belongs to the easy update group, and each geographic attribute item of static / stationary population / economic index belongs to the difficult update group.

  The relationship between geographic attribute items is information indicating the correlation existing between geographic attribute items. In particular, it refers to the relationship between geographic attribute items belonging to the easy update group and geographic attribute items belonging to the difficult update group.

  In addition, the geographic attribute change detection device 1 acquires a text article from the blog article storage device 401, the news article storage device 402, and the like, analyzes the acquired article, and acquires keywords related to the update of the geographic attribute.

  Further, the geographic attribute change detection apparatus 1 obtains a relationship between the geographic attribute items by using the geographic attribute items at a plurality of places and a plurality of points in time and their values. Hereinafter, the geographic attribute change detection apparatus 1 will be described.

  FIG. 2 is a functional block diagram showing the configuration of the geographic attribute change detection device according to the present embodiment. The geographic attribute change detection apparatus 1 shown in FIG. 1 includes a change detection unit 11, a related change detection unit 12, an item relationship accumulation unit 13, an update unit 14, an update required geographic attribute accumulation unit 15, a keyword extraction module 16, and a relationship extraction module. 17. Each unit included in the geographic attribute change detection device 1 may be configured by a computer including an arithmetic processing device, a storage device, and the like, and the processing of each unit may be executed by a program. This program is stored in a storage device included in the geographic attribute change detection device 1, and can be recorded on a recording medium such as a magnetic disk, an optical disk, or a semiconductor memory, or provided through a network. Details of each part will be described below.

  The change detection unit 11 refers to the store database 301, the real estate property database 302, and the like, and when there is a change in the value of the geographic attribute item belonging to the easy update group such as the number of stores and the number of properties, The new value and the name of the place where the change has occurred are acquired and sent to the related change detection unit 12.

  The related change detection unit 12 searches the item relationship storage unit 13 for the relationship between the geographic attribute items including the geographic attribute item for which the change detection unit 11 has detected the change, and detects and updates the geographic attribute item whose change is predicted. The name of the place where the change of the geographic attribute item is predicted is sent to the keyword extraction module 16.

  The item relationship accumulation unit 13 accumulates relationships between geographic attribute items. FIG. 3 shows an example of the relationship between geographic attribute items. In the example shown in the figure, the number of restaurants, which is one of the geographic attribute items, and the “daytime population”, which is one of the geographic attribute items, are stored together with the strength and interpretation of the relationship. A plurality of geographic attribute items may be related to one geographic attribute item. In the example shown in FIG. 3, the “number of restaurants” also has a relationship with another geographic attribute item “number of establishments”. Accumulated.

  The keyword extraction module 16 extracts a keyword related to a place where a change in the geographic attribute item is predicted from an article acquired from the blog article storage device 401, the news article storage device 402, or the like. The keyword extraction module 16 includes an article input unit 161, an article selection unit 162, a place name storage unit 163, and a keyword extraction unit 164.

  The article input unit 161 acquires articles from the blog article storage device 401, the news article storage device 402, and the like.

  The article selection unit 162 selects articles related to the location stored in the location name storage unit 163 from the articles acquired by the article input unit 161.

  The place name accumulation unit 163 accumulates place names for detecting changes in geographic attributes. FIG. 4 shows an example of a place where the place name accumulation unit 163 accumulates. In the example shown in the figure, Marunouchi, Yaesu, Shimbashi, and the like are accumulated as locations for detecting changes in geographical attributes.

  The keyword extraction unit 164 receives an article written about a place where the change of the geographic attribute item detected by the related change detection unit 12 is predicted from the article selection unit 162, extracts a keyword representing the topic of the article, and updates the update unit 14 to send.

  The updating unit 14 stores the location name, the geographic attribute item in which the change is detected, the geographic attribute item that needs to be updated, and the extracted keyword in association with the extracted geographic attribute information in the updating geographic attribute storage unit 15 that needs updating. . FIG. 5 shows an example of update required geographic attribute information accumulated by the update required geographic attribute storage unit 15. The figure shows an example in which the number of restaurants in Marunouchi changes and the daytime population is expected to rise. “Movie theater” and “open” are associated as keywords related to the change in the number of restaurants in Marunouchi.

  The relationship extraction module 17 includes an analysis unit 171 and a past geographic attribute storage unit 172. The analysis unit 171 extracts the relationship between the geographic attribute items using the geographic attribute stored in the past geographic attribute storage unit 172 and stores it in the item relationship storage unit 13. A method for extracting the relationship between geographic attribute items will be described later.

  Next, processing for detecting a geographic attribute item that needs to be updated will be described. FIG. 6 is a flowchart showing a flow of processing for detecting a geographic attribute item that needs to be updated.

  First, the change detection unit 11 refers to the store database 301, the real estate property database 302, and the like, and detects a change in the value of geographic attribute items such as the number of stores and the number of properties (step S11). When a change in the value of the geographic attribute item is detected, the name of the place where the change has occurred, the geographic attribute item, and the value of the new geographic attribute item are sent to the related change detection unit 12.

  Subsequently, the related change detection unit 12 searches for a geographic attribute item whose change is predicted (step S12). The related change detection unit 12 receives the place name, the geographic attribute item, and the value of the new geographic attribute item from the change detection unit 11, and searches the item relationship storage unit 13 for the relationship between the geographic attribute items including the received geographic attribute item. To do. Then, the geographic attribute item whose change is predicted and other information, which are the search results, are sent to the update unit 14. For example, when the relationship between the geographic attribute items shown in FIG. 3 is accumulated in the item relationship accumulation unit 13, if the change detection unit 11 detects an increase in the geographic attribute item “number of restaurants”, the related change detection unit 12. Refers to the item relationship accumulating unit 13 to obtain “daytime population” related to “number of restaurants”. The related change detection unit 12 sends to the update unit 14 the “daytime population” and the strength index value “0.8” as the geographic attribute item that needs to be updated.

  The related change detection unit 12 further searches for a geographic attribute item whose change is predicted for a place that is strongly related to the place where the change has occurred (step S13). When a place that is strongly related to the place where the change has occurred is registered, a geographical attribute item whose change is predicted is also searched for the place that is strongly related, as in step S12. A strongly related place is, for example, a place where public transportation is developed and considered to have a lot of mutual movement. Information on a strongly related place is obtained by the relationship extraction module 17 and stored in the item relationship storage unit 13. A method for extracting a highly related place will be described later.

  Then, when the related change detecting unit 12 sends the place name where the change has occurred to the keyword extracting unit 164 (step S14), the keyword extracting unit 164 receives an article related to the received place name and selects the topic of the article. The keyword to represent is extracted and sent to the update unit 14 (step S15). Details of the keyword extraction process will be described later.

  The update unit 14 associates the received place name, the geographic attribute item whose change is predicted, and the keyword, and stores them in the update-needed geographic attribute storage unit 15 (step S16).

  In this way, the geographic attribute item whose change is predicted, that is, the geographic attribute item that needs to be updated, is extracted and accumulated in the update-needed geographic attribute storage unit 15. By referring to the update-needed geographic attribute information accumulated in the geographic attribute accumulation unit 15 and performing an additional investigation only on the location and geographic attribute items that need to be updated, the geographic attribute can be efficiently kept up-to-date. By keeping the geographic attributes accumulated in the geographic attribute change detection device 1 up-to-date, the effectiveness of the user attributes estimated by the user attribute estimation device 200 can be enhanced, and information distribution suitable for the location can be performed. The administrator can estimate the factors that caused the change in the geographic attribute by referring to the keywords included in the update-required geographic attribute information. It can be determined whether or not to do so.

  Next, processing for extracting keywords will be described. FIG. 7 is a flowchart showing a flow of processing for extracting a keyword.

  The blog article storage device 401 and the news article storage device 402 acquire text articles such as news articles and blog articles from news sites and blog sites published on the network 100 and temporarily store them.

  First, the article input unit 161 acquires the stored articles from the blog article storage device 401 and the news article storage device 402 (step S21).

  Subsequently, the article selection unit 162 selects a place name in the acquired article, and selects an article related to the place accumulated in the place name accumulation unit 163 (step S22). As a method for selecting a place name from the article, the method used in Non-Patent Document 2 may be used.

  Subsequently, the keyword extraction unit 164 extracts keywords having an important meaning in the selected article (step S23). As keywords, general nouns and verbs appearing more than the specified number of times in the same article are extracted, but they appear in the same sentence as the place name, especially when they are included in important positions in the article such as titles. In such a case, weighting such as increasing the evaluation value may be added. As keywords to be extracted, for example, keywords such as “spa”, “opening”, “office building”, “open”, “tower type condominium”, “sales start” can be considered. Note that the method described in Patent Document 1 may be applied.

  The extracted keywords are sent to the update unit 14 and stored in the update-needed geographic attribute storage unit 15 in association with the location attribute and the geographic attribute item whose change is predicted.

  Next, a process for extracting a relationship between geographic attribute items stored in the item relationship storage unit 13 will be described. The relationship between the geographic attribute items is obtained using the history of the geographic attribute items stored in the past geographic attribute storage unit 172 and their values. FIG. 8 is a flowchart showing a flow of processing for extracting a relationship between geographic attribute items.

  First, the analysis unit 171 reads out geographic attribute items and their values at a plurality of places and a plurality of points in time from the past geographic attribute storage unit 172 (step S31). Among the geographic attribute items stored in the past geographic attribute storage unit 172, the values of the geographic attribute items belonging to the easy update group such as the number of restaurants and the number of real estate properties are automatically acquired from the store database 301 and the real estate property database 302. Can be updated.

  Subsequently, the analysis unit 171 uses the values of the geographic attribute items acquired at step S31 at a plurality of locations and a plurality of points in time, between the geographic attribute items belonging to the easy update group and the geographic attribute items belonging to the difficult update group. A relationship is obtained (step S32). Details of how to obtain the relationship between geographic attribute items will be described later.

  When the relationship between the geographic attribute items is obtained, the relationship is stored in the item relationship storage unit 13 (step S33).

  Here, how to obtain the relationship between geographic attribute items will be described.

  First, a method using a correlation coefficient will be described.

  The geographic attribute item and its value are read from the past geographic attribute storage unit 172, the Pearson product-moment correlation coefficient between two of the geographic attribute items is calculated, and the geographic attribute item that belongs to the easy update group and the difficult update group A combination having a large absolute value of the correlation coefficient is found among the geographic attribute items, and the combination of the geographic attribute items is accumulated in the item relationship accumulation unit 13 as a relationship between the geographic attribute items. If there is a change in the value of the geographic attribute item belonging to the easy update group, the value of the geographic attribute item belonging to the difficult update group is also positive if the correlation coefficient is positive, and negative if it is negative. Considered likely to change. The example of the relationship between the geographic attribute items shown in FIG. 3 is obtained by obtaining the relationship between the geographic attribute items using the correlation coefficient.

  Second, a method for considering the influence between places will be described.

  When the geographical attribute item of a certain place changes, the geographical attribute item of another place may change. Here, another location B that is affected by a location A is extracted. When a change occurs in a certain geographic attribute item at location A, it is considered that not only the geographic attribute item closely related to the corresponding geographic attribute item but also the geographic attribute item at location B changes.

  The relationship between places is considered to be related to the number of people who move between each other. In an area where public transportation is developed, the number of people who move from one place to another is considered to depend greatly on the ease of movement between the places (movement cost). For example, Shinjuku-Machida is expected to have more mutual movement than Shibuya-Machida, but it is possible to travel between Shinjuku-Machida without transfer, and frequent express trains operate, but Shibuya-Machida A major reason is that it takes more time and labor to change trains. The travel cost is thought to depend on the time and effort required to travel, the frequency of transportation, the time it takes, and the distance to the nearest transportation station / bus bower.

  Therefore, for example, the movement cost is calculated by the following equation (2).

(Moving cost) = f {a: Traveling time, b: Fare, c: Difficult to use transportation means, d: Number of people getting on and off the nearest public transportation} (2)
a: Travel time, b: Fare can be obtained using a transfer guidance service. The travel time and fare of the most recommended transfer method may be used. c: The difficulty of using the means of transportation may be defined using the distance to the nearest station, the train or bus operation interval, and the like. The distance to the nearest station can be obtained using a transfer guidance service or a map information service. You may rank c from a, and define the sum as a movement cost score. For example, it is assumed that the movement cost is calculated between “Shinagawa” and “Akihabara”. According to a transit service, a: travel time is 15 minutes, rank 3; b: fare is 160 yen, rank 1; c: difficulty of using transportation, 5 minutes except for early morning and late night Since it is less than that, it is ranked 1 and the movement cost score is 5. d: The number of passengers getting on and off the nearest public transport can be obtained by acquiring data published by the transport management company from the database.

  The analysis unit 171 refers to the location name accumulation unit 163, calculates the movement cost between the locations registered in the location name accumulation unit 163, using Equation (2), and obtains a location with much mutual movement. At this time, d: In consideration of the number of passengers in the nearest public transportation system, the number of places to be set may be changed when there is much mutual movement. For example, for the places where the value of d falls in the top 10 places among all the places registered in the place name accumulation unit 163, the places included in the bottom 10 places with the low movement costs are places where there is a lot of mutual movement. For a place where the value of d is small, the places included in the lower three cases where the movement cost is low are assumed to be places where there is much mutual movement.

  Then, the analysis unit 171 accumulates the obtained relationship between the places where the mutual movement is large in the item relationship accumulation unit 13. The relationship between the locations stored in the item relationship storage unit 13 is used in step S13 of FIG.

  As described above, according to the present embodiment, the change detection unit 11 detects a change in the value of a geographic attribute item, and the related change detection unit 12 includes a geographic attribute item including the detected geographic attribute item. By searching the relationship between the items from the item relationship storage unit 13 and detecting a geographic attribute item whose change is predicted, it is possible to grasp the geographic attribute item belonging to the difficult update group that needs to be investigated, and efficiently Geographic attributes can be kept up to date.

  According to the present embodiment, the keyword extraction module 16 extracts a keyword related to a place where a change in geographic attribute items is predicted from articles acquired from the blog article storage device 401, the news article storage device 402, and the like. The update unit 14 stores the location name, the geographic attribute item in which the change is detected, the geographic attribute item that needs to be updated, and the extracted keyword in association with each other in the update required geographic attribute storage unit 15 as update required geographic attribute information. Thus, the administrator who updates the geographic attribute can estimate the factor that caused the change in the geographic attribute by referring to the keyword included in the update required geographic attribute information. It is possible to determine whether additional investigations should be conducted.

  According to the present embodiment, the analysis unit 171 extracts the relationship between the geographic attribute items using the geographic attribute stored in the past geographic attribute storage unit 172 and stores it in the item relationship storage unit 13, so that the change is made. It is possible to detect a geographic attribute item whose change is predicted from the detected geographic attribute item.

DESCRIPTION OF SYMBOLS 1 ... Geographic attribute change detection apparatus 11 ... Change detection part 12 ... Related change detection part 13 ... Item relationship accumulation | storage part 14 ... Update part 15 ... Update geographic attribute accumulation | storage part 16 ... Keyword extraction module 161 ... Article input part 162 ... Article selection Unit 163 ... Location name storage unit 164 ... Keyword extraction unit 17 ... Relationship extraction module 171 ... Analysis unit 172 ... Past geographic attribute storage unit 100 ... Network 200 ... User attribute estimation device 301 ... Store database 302 ... Real estate property database 401 ... Blog article Storage device 402 ... News article storage device

Claims (9)

A step of detecting a change in the value of the first geographic attribute item by referring to a database storing a first geographic attribute item indicating a predetermined attribute related to the place by the change detecting unit;
A second geography having a correlation with the first geographic attribute item and an attribute different from the attribute of the first geographic attribute item by the related change detecting means. Searching for the second geographic attribute item related to the first geographic attribute item detected in the step of detecting the change with reference to a relationship storage means that stores the attribute item in association with each other; and A geographic attribute change detection method characterized by comprising: A step of inputting a text article by an article input means;
Selecting a text article including the place name stored in the place name storage means storing the place names for detecting changes in the values of the first and second geographic attribute items by the article selection means;
Extracting a keyword representing a topic of the text article from a text article including a place name where a change is detected in the first geographic attribute item in the step of detecting the change by a keyword extracting unit;
The update geographic information storage means required by associating the second geographic attribute item retrieved in the step of retrieving the second geographic attribute item with the keyword extracted in the step of extracting the keyword. The method of claim 1, further comprising the step of: A history of the first and second geographic attribute items and their values is read from the past geographic attribute storage means for storing the history of the first and second geographic attribute items and their values by the relationship extracting means, and the first geography is read out. Obtaining a correlation between an attribute item and the second geographic attribute item;
Extracting a relationship between the first geographic attribute item and a second geographic attribute item having a correlation with the first geographic attribute item by the relationship extracting unit, and storing the relationship in the relationship accumulating unit; The geographical attribute change detection method according to claim 1, wherein: The first geography at another location associated with the location where the change of the first geographic attribute item is detected in the step of detecting the change by referring to the relationship accumulation means by the related change detection means. The method for detecting a geographic attribute change according to any one of claims 1 to 3, further comprising a step of searching for the second geographic attribute item related to the attribute item. A first geographic attribute item indicating a predetermined attribute relating to a place, and a second geography having a correlation with the first geographic attribute item and indicating an attribute different from the attribute of the first geographic attribute item Relationship storage means for storing attribute items in association with each other;
Change detecting means for detecting a change in the value of the first geographic attribute item with reference to the database storing the first geographic attribute item;
And a related change detecting means for searching for the second geographic attribute item related to the first geographic attribute item detected by the change detecting means with reference to the relation storage means. Attribute change detection device. A place name accumulating means for accumulating a place name for detecting a change in the value of the first and second geographic attribute items;
An article input means for inputting a text article;
An article selection means for selecting a text article including the place name accumulated in the place name accumulation means;
Keyword extracting means for extracting a keyword representing a topic of the text article from a text article including a place name where the change is detected in the first geographic attribute item in the step of detecting the change;
Update means for associating the second geographic attribute item searched by the related change detection means with the keyword extracted by the keyword extraction means and storing it in the update-needed geographic information storage means. The geographic attribute change detection apparatus according to claim 5. Past geographic attribute storage means for storing a history of first and second geographic attribute items and their values;
A history of the first and second geographic attribute items and their values is read from the past geographic attribute storage means, a correlation between the first geographic attribute item and the second geographic attribute item is obtained, and the first geographic attribute item is obtained. A relationship extracting unit that extracts a relationship between a geographic attribute item and a second geographic attribute item having a correlation with the first geographic attribute item, and stores the relationship in the relationship accumulating unit. The geographic attribute change detection apparatus according to claim 5 or 6. The related change detection means refers to the relation storage means, and the first change related to the first geographic attribute item in another location associated with the place where the change of the first geographic attribute item is detected. The geographic attribute change detecting device according to claim 5, wherein two geographic attribute items are searched.   A geographic attribute change detection program for causing a computer to execute the geographic attribute change detection method according to claim 1.

Images