JP5464976B2 - Information processing apparatus, information processing method, and information processing program - Google Patents

Description

  The present invention relates to an information processing apparatus having a function for selecting an appropriate combination of a place name keyword and an address keyword to be associated with event information acquired by database search.

  Conventionally, various approaches (digital archiving projects, etc.) have been carried out from the large organizational level to the individual level for collecting historical information. However, historical information (videos, images, literature, etc.) collected by various projects is only used within individual projects and is rarely used between multiple projects. Currently. On the other hand, information collected in this way (also referred to as historical event information) is often published so that it can be viewed on the Internet as well as presented at academic conferences and the like.

  Thus, techniques (applications) for collecting historical event information using Internet search and mapping the information on a map have been proposed. In this conventional application, when a user inputs a predetermined keyword, historical event information associated with the keyword is collected using an Internet search, and place name information is extracted from the historical event information (part of the sentence) Mapping on the map is performed based on the place name information (see, for example, Patent Document 1).

JP 2009-087140 A

  However, in the conventional application, it is possible to extract the place name from the sentence part included in the historical event information, but the extracted place name indicates the place name of the historical event information (place name to be associated with the historical event information). It is difficult to select an appropriate place name from among them, especially when multiple place names are included in the text. .

  Furthermore, in an application that maps historical event information on a map, it is necessary to select an appropriate combination of the place name and address in order to perform an appropriate mapping. Was not.

  The present invention has been made in view of the above problems, and an object thereof is to provide an information processing apparatus that can determine the appropriateness of a combination of a place name keyword and an address keyword to be associated with event information.

  The information processing apparatus of the present invention uses a database search to extract an event information collection unit that collects event information associated with a predetermined person name keyword, and extracts a place name keyword from the event text included in the collected event information. A place name extraction unit; an event name extraction unit that extracts an event name keyword from the event sentence from which the place name keyword has been extracted; and a database search, and the number of event sentences including the person name keyword and the event name keyword A place name co-occurrence degree calculating unit that calculates a place name co-occurrence degree indicating a degree of the number of event sentences including the place name keyword, and an address keyword related to the place name keyword is obtained for the place name keyword for which the place name co-occurrence degree is calculated. Using the address acquisition unit and database search An address co-occurrence degree calculating unit for calculating an address co-occurrence degree indicating a degree of the number of event sentences including the address keyword among the number of event sentences including the person name keyword and the event name keyword, the place name co-occurrence degree and the address An appropriateness calculating unit that calculates an appropriateness of a combination of a place name keyword and an address keyword to be associated with the event information based on the co-occurrence degree;

  As a result, when event information is collected using personal name keywords by database search, and place name keywords are extracted from the event text included in the event information, the place name co-occurrence and address co-occurrence calculated using database search are collected. Based on the degree of occurrence, the appropriateness of the combination of the place name keyword and the address keyword to be associated with the event information can be obtained. Therefore, for example, when applying the event information to an application that maps and displays the event information on the map based on the place name keyword and the address keyword, it is easy to appropriately combine the place name keyword and the address keyword to be associated with the event information. You can choose to.

  Further, the information processing apparatus according to the present invention determines that the place name keyword is inappropriate as the place name keyword to be associated with the event information when the calculated place name co-occurrence degree is smaller than a predetermined threshold. In addition, an inappropriateness determination unit that determines not to acquire the address keyword and calculate the address co-occurrence degree may be provided.

  As a result, when the place name co-occurrence value calculated for the place name keyword is small, it is determined that the place name keyword to be associated with the event information is inappropriate and the address keyword acquisition process and the address co-occurrence degree calculation process are performed. Since it is not performed, unnecessary processing can be reduced and the calculation load can be reduced.

  In the information processing apparatus of the present invention, if the event name extraction unit cannot extract the event name keyword from the event sentence, the event name extraction unit extracts a rival person name keyword of the person name keyword from the event sentence, The place name co-occurrence degree calculation unit and the address co-occurrence degree calculation unit may use the rival person name keyword instead of the event name keyword.

  Thereby, when an event name keyword cannot be extracted from an event sentence, a rival person name keyword can be used instead.

  In the information processing apparatus of the present invention, the address acquisition unit may acquire an address keyword related to a current place name notation corresponding to the old place name notation when the place name keyword is an old place name notation. Good.

  Thereby, when a place name keyword is an old place name notation, an address keyword related to the current place name notation corresponding to the old place name notation can be acquired.

  The information processing method of the present invention collects event information associated with a predetermined person name keyword using database search, extracts a place name keyword from event sentences included in the collected event information, and selects the place name keyword. An event name keyword is extracted from the extracted event sentence, and the name of the place name indicating the degree of the number of event sentences including the place name keyword out of the number of event sentences including the person name keyword and the event name keyword is determined using database search. An event text including the person name keyword and the event name keyword is obtained by calculating an occurrence degree, obtaining an address keyword related to the place name keyword for the place name keyword for which the place name co-occurrence degree is calculated, and using a database search The address keyword out of the number Calculating the degree of address co-occurrence indicating the degree of the number of event sentences included, and determining the appropriateness of the combination of the place name keyword and the address keyword to be associated with the event information based on the place name co-occurrence degree and the address co-occurrence degree calculate.

  Also by this method, the same effect as the above apparatus can be obtained. That is, for the event information acquired by the database search, the appropriateness of the combination of the place name keyword and the address keyword to be associated with the event information can be obtained.

  The information processing program of the present invention uses a database search in a computer to collect event information associated with a predetermined person name keyword and extract a place name keyword from the event text included in the collected event information. An event including the place name keyword out of the number of event sentences including the person name keyword and the event name keyword using a database search, and a process for extracting the event name keyword from the event sentence from which the place name keyword is extracted A process for calculating a place name co-occurrence degree indicating the degree of the number of sentences, a process for obtaining an address keyword related to the place name keyword, and a database search for the place name keyword for which the place name co-occurrence degree is calculated , The name keyword and the a A process for calculating an address co-occurrence degree indicating a degree occupied by the number of event sentences including the address keyword among the number of event sentences including the event name keyword, and the event based on the place name co-occurrence degree and the address co-occurrence degree And a process of calculating the appropriateness of the combination of the place name keyword and the address keyword to be associated with the information.

  Also by this program, the same effect as the above apparatus can be obtained. That is, for the event information acquired by the database search, the appropriateness of the combination of the place name keyword and the address keyword to be associated with the event information can be obtained.

  According to the present invention, by providing a place name co-occurrence degree calculation unit, an address co-occurrence degree calculation unit, and an appropriateness degree calculation unit, based on the place name co-occurrence degree and the address co-occurrence degree calculated using database search, The appropriateness of the combination of the place name keyword and the address keyword to be associated with the event information can be obtained.

The block diagram which shows the structure of the information processing apparatus in embodiment of this invention The figure which shows an example of the process which extracts a place name keyword and an event name keyword from event information (event text) The figure which shows an example of the address keyword acquired from the place name keyword The figure which shows an example of the place name co-occurrence degree calculated about each place name keyword The figure which shows an example of the address co-occurrence degree calculated about each of the address keyword The figure which shows an example of the appropriateness of the combination of a place name keyword and an address keyword The flowchart which shows the flow of operation | movement of the information processing apparatus in embodiment of this invention.

  Hereinafter, an information processing apparatus according to an embodiment of the present invention will be described with reference to the drawings. In the present embodiment, a case of an information processing apparatus applied to an application or the like that maps and displays historical event information on a map based on a place name keyword and an address keyword is illustrated. Such applications include, for example, educational software and tourism software that maps and displays historical content on a map, vehicle navigation software that uses old maps, and automatic history guides that read historical guides about historic sites near vehicles. Software etc. are included.

  The information processing apparatus according to the present embodiment has a function of determining the appropriateness of a combination of a place name keyword and an address keyword for the historical event information so that the historical event information can be appropriately used in such an application. Such a function is realized by a program stored in the HDD or memory of the information processing apparatus.

  A configuration of an information processing apparatus according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the information processing apparatus 1 according to the present embodiment. As shown in FIG. 1, the information processing apparatus 1 includes an event information collection unit 4 that collects historical event information from various databases 3 connected to a network 2 such as the Internet. The event information collection unit 4 includes historical event information (content information such as videos, images, and documents) associated with a personal name keyword (for example, Nobunaga Oda) input by a user from a personal name input unit 5 (for example, a keyboard). Are collected using internet search. The historical event information collected by the event information collection unit 4 is stored in a storage unit 6 (for example, a large-capacity hard disk device) provided in the information processing apparatus 1.

  When the information processing apparatus 1 has a function for automatically collecting personal name keywords, the event information collection unit 4 uses not only the personal name keyword input by the user but also the personal name keyword acquired by automatic collection. Internet search may be performed. In addition, the event information collection unit 4 may have a crawling function, and historical event information collected by performing an Internet search in advance may be cached in the storage unit 6. In that case, the event information collection unit 4 executes a search process on the historical event information cached in the storage unit 6, so that an extremely high-speed search is possible.

  The information processing apparatus 1 also includes a place name extraction unit 7 that extracts place name keywords from sentences (event sentences) included in historical event information collected by the event information collection unit 4, and an event name keyword from the event sentences from which the place name keywords are extracted. An event name extraction unit 8 is provided. These place name extraction unit 7 and event name extraction unit 8 have a function of extracting a keyword from an event sentence by, for example, morphological analysis.

  FIG. 2 is a diagram illustrating an example of processing for extracting a place name keyword and an event name keyword from event information (event text). For example, it is assumed that historical event information on “Hieizan Yaki Kage” was obtained as a result of a database search using a personal name keyword “Oda Nobunaga”. As shown in FIG. 2, this historical event information includes an event sentence and an event image relating to “Hieizan baked down”. In this case, the place name extraction unit 7 can extract place name keywords “loss”, “Sakamoto”, and “Mt. Hiei” from this event sentence by morphological analysis. In addition, the event name extraction unit 8 can extract an event name keyword “Make Hieiyama” from this event sentence by morphological analysis.

  Depending on historical event information, the event text may not be included in the event text. In this case, the event name extraction unit 8 cannot extract the event name keyword from the event text. However, instead, the event name extraction unit 8 can extract the rival name keyword from the event text. The rival person name keyword is a person name (for example, Mitsuhide Akechi) of a person who is regarded as a historical rival of the person name keyword (for example, Nobunaga Oda) used when searching for the historical event information. This rival person name keyword can be used in place of the event name keyword in subsequent processing (processing such as calculation of place name co-occurrence degree and address co-occurrence degree described later).

  The information processing apparatus 1 further includes an address acquisition unit 9 that acquires an address keyword related to the place name keyword for the place name keyword extracted from the historical event information. The address acquisition unit 9 has a function of acquiring an address keyword related to the place name keyword, for example, by geocoding or the like.

  FIG. 3 is a diagram illustrating an example of an address keyword acquired from a place name keyword. As shown in FIG. 3, for example, when geocoding is performed on a place name keyword “LOSS”, address keywords “California” and “Los Angeles” can be acquired. In addition, if geocoding is applied to the place name keyword "Sakamoto", the address keywords "Kumamoto Prefecture", "Yatsushiro City", "Sakamoto" and the address keywords "Shiga Prefecture", "Otsu City", "Sakamoto" can be obtained. it can. Further, if geocoding is applied to the place name keyword “Mt. Hiei”, address keywords “Shiga Prefecture”, “Otsu City”, and “Sakamoto” can be acquired.

  The address acquisition unit 9 can also deal with the case where the place name keyword is an old place name notation, that is, when the place name keyword is an old place name notation, the current place name notation corresponding to the old place name notation is used. It has a function to acquire address keywords related to. For example, for place name keywords of old place names such as “Omi” and “Edo”, refer to place name notation database (not shown), etc., and related to the current place name notation of “Shiga Prefecture” or “Tokyo” Address keywords can be acquired.

  In addition, the information processing apparatus 1 includes an arithmetic processing unit 10 configured by a CPU, a microcomputer, and the like. The arithmetic processing unit 10 includes a place name co-occurrence degree calculation unit 11, an inappropriateness determination unit 12, and an address co-occurrence degree. A calculation unit 13 and an appropriateness calculation unit 14 are provided.

The place name co-occurrence degree calculation unit 11 uses a database search to calculate a place name co-occurrence degree indicating the degree of the number of event sentences including the place name keyword among the number of event sentences including the person name keyword and the event name keyword. It has. For example, the place name co-occurrence degree calculation unit 11 can calculate the place name co-occurrence degree using the following Equation 1.
P (T _i | H, E) = N _{H, E} (T _i ) / N _{H, E} (Equation 1)
Here, P (T _i | H, E) is a place name co-occurrence degree for the place name T _i . N _{H and E} are the number of sentences including the person name H and the event name E obtained by database search, and N _{H and E} (T _i ) are sentences including the place name T _i out of the number of sentences N _{H and E.} Is a number.

  FIG. 4 is a diagram illustrating an example of the place name co-occurrence degree calculated by the place name co-occurrence degree calculating unit 11. As shown in FIG. 4, for example, the place name co-occurrence degree for the place name keyword “loss” is calculated as “0.0192”. The place name co-occurrence degree for the place name keyword “Sakamoto” is calculated as “0.3728”. Also, the place name co-occurrence degree for the place name keyword “Hieizan” is calculated as “0.8065”.

  When the place name co-occurrence calculated as described above is smaller than a predetermined threshold (for example, 0.1000), the inappropriateness determination unit 12 is inappropriate as a place name keyword to be associated with historical event information. For the place name keyword, it is determined that the acquisition of the address keyword and the calculation of the address co-occurrence degree are not performed. For example, in the case of FIG. 4, the place name co-occurrence of the place name keyword “loss” is “0.0192”, which is smaller than the threshold value “0.1000”. Processing such as acquisition of address keywords and calculation of address co-occurrence is not performed.

The address co-occurrence degree calculation unit 13 uses a database search to calculate an address co-occurrence degree indicating the degree of the event sentence number including the address keyword among the number of event sentences including the person name keyword and the event name keyword. It has. For example, the address co-occurrence degree calculation unit 13 can calculate the address co-occurrence degree using Equation 2 below.
P (A _j | H, E) = N _{H, E} (A _j ) / N _{H, E} (Expression 2)
Here, P (A _j | H, E) is an address co-occurrence degree for the address A _j . N _{H and E} are the number of sentences including the person name H and the event name E obtained by database search, and N _{H and E} (A _j ) are sentences including the address A _j among the number of sentences N _{H and E.} Is a number.

  FIG. 5 is a diagram illustrating an example of the address co-occurrence degree calculated by the address co-occurrence degree calculation unit 13. As shown in FIG. 5, for example, the address co-occurrence degree for the address keyword “Shiga Prefecture” is calculated as “0.2753”. The address co-occurrence degree for the address keyword “Otsu City” is calculated as “0.2674”. The address co-occurrence degree for the address keyword “Sakamoto” is calculated as “0.3728”. The address co-occurrence degree for the address keyword “Kumamoto Prefecture” is calculated as “0.0412”. Also, the address co-occurrence degree for the address keyword “Yatsushiro City” is calculated as “0.0014”.

  Note that the address co-occurrence of the address keyword “California” and “Los Angeles” does not need to be calculated for the place name keyword “Loss”. The degree is calculated as “0.0014”. The address co-occurrence degree for the address keyword “Los Angeles” is calculated as “0.0087”.

The appropriateness calculation unit 14 has a function of calculating the appropriateness of the combination of the place name keyword and the address keyword to be associated with the historical event information based on the place name co-occurrence degree and the address co-occurrence degree. For example, the appropriateness calculation unit 14 can calculate the appropriateness S using the following Equation 3.
S = P (T _i | H, E) * P (A _j | H, E) (Equation 3)

  FIG. 6 is a diagram illustrating an example of the appropriateness calculated by the appropriateness calculating unit 14. As shown in FIG. 6, the place name co-occurrence calculated for place name keywords such as “Sakamoto” and “Hieizan” and the address co-occurrence calculated for address keywords such as “Shiga Prefecture” and “Otsu City” By accumulating each, the appropriateness of the combination of each place name keyword and address keyword is calculated. In FIG. 6, combinations of place name keywords and address keywords are sorted in descending order of appropriateness. In this example, the place keyword “Mt. Hiei” and the address keyword “Shiga Prefecture, Otsu City, Sakamoto” have the highest value “0.3006”.

  Although not specifically shown here, when an address keyword is acquired from a place name keyword using geocoding, the “latitude / longitude” corresponding to the address keyword, “accuracy” of the address keyword, etc. It is also possible to acquire information together. The accuracy is a numerical value indicating the fineness and accuracy of the address keyword. For example, the accuracy of the address keyword “Shiga Prefecture” is “4”, the accuracy of the address keyword “Shiga Prefecture, Otsu City” is “3”, and the address keyword “Shiga Prefecture, Otsu City, Sakamoto” Accuracy is “2”.

  In the example of FIG. 6, as a result of sorting in descending order of appropriateness, combinations of address keywords “Shiga Prefecture, Otsu City, Sakamoto” with the smallest accuracy are sorted in the highest order. For example, in an application such as vehicle navigation software using an old map, it is desirable to sort the address keywords with a low accuracy in this way so that they are higher. However, it goes without saying that, depending on the type and content of the application, the address keywords with large accuracy may be sorted so that they can be used in the application.

  The operation of the information processing apparatus 1 configured as described above will be described with reference to the drawings. Here, the characteristic operation of the present invention, that is, the flow of the operation for calculating the appropriateness of the combination of the place name keyword and the address keyword to be associated with the event information will be described with reference to the flowchart of FIG.

  As shown in FIG. 7, when the degree of fitness is calculated using the information processing apparatus 1 of the present embodiment, first, when a personal name keyword such as “Oda Nobunaga” is input from the personal name input unit 5 (S <b> 1). The event information collection unit 4 collects historical event information associated with the personal name keyword (S2). For example, when historical event information as shown in FIG. 2 is collected, place name keywords such as “Loss”, “Sakamoto”, and “Hieizan” are extracted from the historical event information by morphological analysis (S3). Event name keywords such as "are extracted (S4). Then, as shown in FIG. 4, the place name co-occurrence degree is calculated for each place name keyword using the above-described formula 1 (S5).

  Next, it is determined whether or not the place name co-occurrence calculated for each place name keyword is greater than a predetermined threshold (S6). For example, for place name keywords such as “Sakamoto” and “Mt. Hiei” where place name co-occurrence is determined to be greater than the threshold value, as shown in FIG. Address keywords such as “city” are acquired (S7). And as shown in FIG. 5, the address co-occurrence degree is calculated about each address keyword using the above-mentioned Formula 2 (S8).

  Finally, from the place name co-occurrence degree and the address co-occurrence degree calculated in this way, as shown in FIG. 6, the suitability of the combination of the place name keyword and the address keyword is calculated using the above-described equation 3. (S9). In addition, in the above determination step (S6), for the place name keyword determined to have a place name co-occurrence degree smaller than the threshold, such as “loss”, for example, the subsequent processing (S7 to S9) is not performed. End the operation.

  According to the information processing apparatus 1 according to the embodiment of the present invention, by providing the place name co-occurrence degree calculation unit 11, the address co-occurrence degree calculation unit 13, and the appropriateness degree calculation unit 14, the database search is used. Based on the calculated place name co-occurrence degree and address co-occurrence degree, the appropriateness of the combination of the place name keyword and the address keyword to be associated with the event information can be obtained.

  That is, in the present embodiment, historical event information is collected using a personal name keyword (for example, Nobunaga Oda) by database search, and a place name keyword (for example, Mt. Hiei) is extracted from the event text included in the historical event information. When this is done, based on the place name co-occurrence degree and the address co-occurrence degree calculated using the database search, a place name keyword (for example, Mt. Hiei) to be associated with the historical event information and a combination of address keywords (for example, Shiga) Appropriateness of Sakamoto, Otsu City, etc.). Thus, for example, when applied to an application that maps historical event information on a map based on place name keywords and address keywords, an appropriate combination of place name keywords and address keywords to be associated with the historical event information Can be selected easily.

  In the present embodiment, if the place name co-occurrence value calculated for the place name keyword (for example, loss) is small, it is determined that the place name keyword to be associated with the historical event information is inappropriate as the address keyword. Acquisition processing and address co-occurrence degree calculation processing are not performed, unnecessary processing can be reduced, and calculation load can be reduced.

  Further, in the present embodiment, when an event name keyword (for example, the Honnoji ware eradication) cannot be extracted from the event text, a rival person name keyword (for example, Mitsuhide Akechi) can be used instead.

  In this embodiment, when the place name keyword is an old place name notation, an address keyword related to the current place name notation (for example, Shiga Prefecture) corresponding to the old place name notation (for example, Omi) is acquired. Can do.

  The embodiments of the present invention have been described above by way of example, but the scope of the present invention is not limited to these embodiments, and can be changed or modified according to the purpose within the scope of the claims. is there.

  As described above, the information processing apparatus according to the present invention has an effect that the appropriateness of the combination of the place name keyword and the address keyword to be associated with the event information can be obtained, and the event information is converted into the place name keyword and the address keyword. The present invention is useful because it is applied to an application that maps and displays it on a map based on it.

DESCRIPTION OF SYMBOLS 1 Information processing apparatus 2 Network 3 Database 4 Event information collection part 5 Person name input part 6 Storage part 7 Place name extraction part 8 Event name extraction part 9 Address acquisition part 10 Arithmetic processing part 11 Place name co-occurrence degree calculation part 12 Inappropriate judgment part 13 Address co-occurrence calculation unit 14 Adequacy calculation unit

Claims (5)

An event information collection unit that collects event information associated with a predetermined person name keyword using a database search;
A place name extraction unit that extracts place name keywords from the event text included in the collected event information;
An event name extraction unit for extracting an event name keyword from the event sentence from which the place name keyword has been extracted;
Using a database search, a place name co-occurrence degree calculating unit that calculates a place name co-occurrence degree indicating the degree of the number of event sentences including the place name keyword among the number of event sentences including the person name keyword and the event name keyword;
For the place name keyword for which the place name co-occurrence degree is calculated, an address acquisition unit that acquires an address keyword related to the place name keyword;
Using a database search, an address co-occurrence degree calculating unit that calculates an address co-occurrence degree indicating the degree of the event sentence number including the address keyword among the number of event sentences including the person name keyword and the event name keyword;
Based on the place name co-occurrence degree and the address co-occurrence degree, the appropriateness calculation unit that calculates the appropriateness of the combination of the place name keyword and the address keyword to be associated with the event information according to the following equation :
An information processing apparatus comprising:
S = P (T _i | P, E) * P (A _j | P, E) (formula)
Here, S is the suitability degree, P (T _i | P, E) is the place name co-occurrence degree, and P (A _j | P, E) is the address co-occurrence degree.   For the place name keyword, if the calculated place name co-occurrence is smaller than a predetermined threshold, it is determined that the place name keyword should be associated with the event information, and the acquisition of the address keyword and the address share are determined. The information processing apparatus according to claim 1, further comprising an inappropriateness determination unit that determines not to calculate the degree of occurrence. The information according to claim 1 or 2 , wherein the address acquisition unit acquires an address keyword related to a current place name notation corresponding to the old place name notation when the place name keyword is an old place name notation. Processing equipment. Use database search to collect event information associated with a given person keyword,
The place name keyword is extracted from the event text included in the collected event information,
The event name keyword is extracted from the event sentence from which the place name keyword is extracted,
Using a database search, a place name co-occurrence degree indicating the degree of the number of event sentences including the place name keyword out of the number of event sentences including the person name keyword and the event name keyword is calculated,
For the place name keyword for which the place name co-occurrence degree is calculated, an address keyword related to the place name keyword is acquired,
Using a database search, calculate the degree of address co-occurrence indicating the degree of the number of event sentences including the address keyword out of the number of event sentences including the person name keyword and the event name keyword,
An information processing method comprising: calculating an appropriateness of a combination of a place name keyword and an address keyword to be associated with the event information based on the place name co-occurrence degree and the address co-occurrence degree according to the following expression .
S = P (T _i | P, E) * P (A _j | P, E) (formula)
Here, S is the suitability degree, P (T _i | P, E) is the place name co-occurrence degree, and P (A _j | P, E) is the address co-occurrence degree. On the computer,
Using database search to collect event information associated with a given person keyword,
A process of extracting place name keywords from the event text included in the collected event information;
A process of extracting an event name keyword from the event sentence from which the place name keyword has been extracted;
Using a database search, calculating a place name co-occurrence degree indicating a degree occupied by the number of event sentences including the place name keyword among the number of event sentences including the person name keyword and the event name keyword;
For the place name keyword for which the place name co-occurrence degree is calculated, a process for obtaining an address keyword related to the place name keyword;
A process of calculating an address co-occurrence degree indicating a degree occupied by the number of event sentences including the address keyword among the number of event sentences including the person name keyword and the event name keyword using a database search;
According to the following formula , based on the place name co-occurrence degree and the address co-occurrence degree, a process for calculating the appropriateness of the combination of the place name keyword and the address keyword to be associated with the event information;
An information processing program for executing
S = P (T _i | P, E) * P (A _j | P, E) (formula)
Here, S is the suitability degree, P (T _i | P, E) is the place name co-occurrence degree, and P (A _j | P, E) is the address co-occurrence degree.