JP2018180638A - Name identification support device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a name identification supporting device which can more precisely determine the same name with the same place information.SOLUTION: A server 10 according to an embodiment includes: a POI information pair acquisition part 12 for acquiring a pair of POI information indicating the same POI from plural pieces of POI information each including category information indicating the category of a POI; and a category pair acquisition part 13 for acquiring a category pair as a pair of category information showing the same category on the basis of the combination of category information included in each piece of POI information included in the pair acquired by the POI information pair acquisition part 12.SELECTED DRAWING: Figure 1

Description

  One aspect of the present invention relates to a name identification support device.

  Conventionally, when using location information (for example, information related to POI (Point Of Interest)) provided by, for example, a plurality of different data sources (for example, information service providers), merging these location information ) Has been done. For example, in Patent Document 1, when the difference between coordinate data associated with each of two pieces of place information (facility information) (that is, the distance between places) falls within a threshold range, each of the two pieces of place information There is disclosed a method of determining that indicates the same place.

JP, 2010-086399, A

  According to the method disclosed in Patent Document 1, the identity (whether or not to indicate the same place) of the place information can be determined based on the coordinate data associated with each place information, but the place information is more accurate. There is a need for a mechanism to merge (merge).

  Therefore, one aspect of the present invention is to provide a name identification support device for more accurately determining the identity of location information.

  A location identification support device according to an aspect of the present invention includes a location information pair acquisition unit that acquires a pair of location information indicating the same location from among a plurality of location information each including category information indicating a category of location; A category pair acquisition unit for acquiring a category pair which is a pair of category information indicating the same category based on a combination of category information included in each of the location information included in the pair acquired by the information pair acquisition unit; Prepare.

  In the name identification support device according to one aspect of the present invention, a pair of location information indicating the same location is acquired from among the plurality of location information. Then, based on a combination of category information included in the pair, a category pair which is a pair of category information indicating the same category is acquired. By combining the identity of places and the identities of categories in this manner, category pairs indicating the same category can be generated with high accuracy. Such category pairs can be used to accurately determine the identity of location information. Therefore, according to the name identification support device, it is possible to more accurately determine the identity of location information.

  According to one aspect of the present invention, it is possible to provide a name identification support device for more accurately determining the identity of location information.

It is a figure which shows the function structure of the server containing the name identification assistance apparatus which concerns on one Embodiment. It is a figure which shows an example of a POI information pair candidate management table. It is a figure which shows an example of a POI information pair management table. It is a figure which shows an example of a category pair candidate management table. It is a figure which shows an example of a comparison object pair management table. It is a figure which shows an example of a comparison object pair management table. It is a flowchart which shows operation | movement of the server of FIG. It is a figure which shows the hardware constitutions of the server of FIG.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the attached drawings. In the description of the drawings, the same or corresponding elements will be denoted by the same reference symbols, without redundant description.

  FIG. 1 is a diagram showing a functional configuration of a server 10 including a name identification support device according to an embodiment. The server 10 is an apparatus that executes name identification of a plurality of POI information (place information). The plurality of POI information is, for example, a set of POI information provided by a plurality of different data sources (for example, information service providers).

  The “POI information” is information on a point of interest (POI) indicating a specific place (for example, a facility or the like). Examples of POI include stores, leisure facilities, and tourist facilities. The POI information is information indicating the name of the POI (for example, a character string), positional information indicating the position coordinates associated with the POI (for example, latitude and longitude), and a category of the POI (for example, genre of facilities such as "eater") May include category information (eg, a character string) indicating.

  The “name identification of POI information” means that POI information indicating the same POI is merged to eliminate duplication of information. For example, in POI information provided by a plurality of different data sources, there may be duplicate POI information related to the same POI. The detailed information such as the POI name, position information, and category information included in such POI information may or may not match each other. Discrepancies in detailed information are attributable to differences in information held by each data source. For this reason, in order to appropriately combine a plurality of POI information, it is specified as a POI information (POI information indicating the same POI) which mutually overlaps a pair in which detailed information (name, position information, category information, etc.) completely matches It is not enough to do just that. That is, since some of the detailed information may indicate the same POI even for different pairs, it is possible to appropriately determine the identity (whether or not the same POI is indicated) between the POI information for such pairs as well. It is required to

  The server 10 acquires category pairs, which are pairs of category information indicating the same category, as processing for supporting such name identification of POI information, and uses the category pairs to compare with the first POI information to be compared. The determination (identification determination) of the identity of the second POI information is executed. As a specific example of the category pair, a category indicating a same category (in this case, a category indicating a restaurant providing a sushi) but different character strings, such as category information "sushi store" and category information "sushi" There is a pair of information. Note that since category information having the same character string naturally indicates the same category, this embodiment mainly determines whether pairs of category information having different character strings indicate the same category. It is assumed.

  Each functional element of the server 10 will be described with reference to FIG. As shown in the figure, the server 10 includes a storage unit 11, a POI information pair acquisition unit 12 (place information pair acquisition unit), a category pair acquisition unit 13, a comparison target acquisition unit 14, a determination unit 15, and And. The POI information pair acquisition unit 12 and the category pair acquisition unit 13 are functional elements for acquiring a category pair. The comparison target acquisition unit 14 and the determination unit 15 are functional elements for performing name identification determination on the first POI information and the second POI information to be compared.

  The storage unit 11 stores various types of information necessary for the processing of the server 10. In the present embodiment, the storage unit 11 includes a POI information pair candidate management table (see FIG. 2), a POI information pair management table (see FIG. 3), a category pair candidate management table (see FIG. 4), and a comparison target pair management table. (See FIGS. 5 and 6) Various information stored in the storage unit 11 can be accessed from the POI information pair acquisition unit 12, the category pair acquisition unit 13, the comparison target acquisition unit 14, and the determination unit 15.

  The POI information pair acquisition unit 12 acquires a pair of POI information indicating the same POI from among a plurality of POI information. For example, the POI information pair acquiring unit 12 receives two POI information (POI information pair candidates) to be compared as inputs, and determines whether the two POI information indicate the same POI based on a predetermined determination criterion It is determined whether or not. As an example, the POI information pair candidate is stored in advance in the POI information pair candidate management table. For example, the POI information pair candidate management table stores POI information pair candidates corresponding to all combinations obtained by extracting arbitrary two pieces of POI information from among a plurality of POI information provided by a plurality of data sources. It is done.

  FIG. 2 is a diagram showing an example of the POI information pair candidate management table. In this example, the POI information pair candidate management table stores a plurality of POI information pair candidates. The POI information pair candidate management table stores two pieces of POI information (POI information A and POI information B) to be compared as information (information for one record) indicating one POI information pair candidate. POI information A includes POI name A, category information A, and position information (latitude A and longitude A). POI information B includes POI name B, category information B, and position information (latitude B and longitude B). The alphabets “A” to “E” in the “Name A” column or “Name B” column in FIG. 2 each represent a character string consisting of one or more characters, and the same alphabet represents the same character string. There is. Further, the values stored in the “dist” column and the “lcsr” column of the POI information pair candidate management table are values calculated by the processing of the POI information pair acquisition unit 12 described later.

In the present embodiment, the POI information pair acquiring unit 12 satisfies, in the POI information pair candidate (POI information A and POI information B), a first predetermined condition (details will be described later) for the name and position of the POI. It is determined whether or not. Then, based on the result of the determination, the POI information pair acquisition unit 12 determines whether or not the POI information pair candidate indicates the same POI. First, for each POI information pair candidate, the POI information pair acquiring unit 12 generates POI information based on position information (latitude A and longitude A) of POI information A and position information (latitude B and longitude B) of POI information B. A distance dist, which is a linear distance between the POI indicated by A and the POI indicated by the POI information B, is calculated. For example, assuming that the position coordinate of POI information A is (x1, y1) and the position coordinate of POI information B is (x2, y2), POI information pair acquiring unit 12 calculates distance dist by the following (formula 1) can do. In the following (Formula 1), r represents the equatorial radius (6378.137 km).
(Expression 1): dist = r × cos−1 (sin (y1) × sin (y2) + cos (y1) × cos (y2) × cos (x2-x1))

  The POI information pair acquiring unit 12 stores the distance dist thus calculated in the “dist” column of the POI information pair candidate management table. In the example of FIG. 2, the distance “7.81 (m)” is calculated for the POI information pair candidate in the first row.

Further, the POI information pair acquisition unit 12 calculates the longest common subsequence ratio lcsr of the name A and the name B for each POI information pair candidate. First, the POI information pair acquisition unit 12 extracts the longest common subsequence of the name A and the name B. The common substring is a substring in which the same elements appear successively in two strings. The longest common subsequence is the longest (highest number of characters) subsequence among the possible common subsequences. The longest common part, which is the length (number of characters) of the longest common substring of two strings X and Y, as shown in the following (Expression 2), the longest common substring ratio lcsr of the two strings X and Y: It is a value obtained by dividing the column length (LLCS: Length of Longest Common Subsequence) by the string length (number of characters) of the longer one of the two strings X and Y.
(Expression 2): lcsr = LLCS (X, Y) / max (L (X), L (Y))
Here, L (X) represents the string length of the string X. max (x, y) represents the larger value of x and y.

  As a specific example, consider a case where one character string X is "Sushi XXX mall Totsuka store" and the other character string Y is "XXX mall Totsuka store". In this case, the longest common partial string of the character string X and the character string Y is “XXX mall Totsuka store”, and the LLCS (X, Y) is 9. Further, L (X) is 11, and L (Y) is 9, so max (L (X), L (Y)) is 11. Therefore, in this example, lcsr is 0.818 (= 9/11).

  The POI information pair acquiring unit 12 calculates the longest common subsequence ratio lcsr between the name A and the name B by the above (formula 2), and stores the calculated lcsr in the “lcsr” column of the POI information pair candidate management table . In the example of FIG. 2, lcsr “0.800” is calculated for the POI information pair candidate in the first row.

Subsequently, the POI information pair acquiring unit 12 determines whether the POI information pair candidate indicates the same POI, based on the distance dist and the longest common subsequence ratio lcsr calculated as described above. For example, with respect to the POI indicated by the POI information A and the POI indicated by the POI information B, when the distance dist is small and the names completely match (namely, lcsr = 1), the POI information A and the POI information B have the same POI. Is likely to indicate. Therefore, in the case where the POI information pair acquiring unit 12 satisfies, for example, the first condition (the condition for the POI name (lcsr = 1) and the condition for the POI position (dist ≦ 30) defined in advance as follows: Alternatively, it may be determined that the POI information pair candidates indicate the same POI.
(First condition)
・ Lcsr = 1
・ Dist ≦ 30

  Here, in order to obtain category pairs to be described later with high accuracy, it is important to obtain (extract) a correct POI information pair. For this reason, the first condition is set to a relatively severe condition in order to prevent that POI information which does not actually indicate the same POI is erroneously extracted as a POI information pair (false extraction). There is.

  The POI information pair acquiring unit 12 acquires (extracts) POI information pair candidates determined to satisfy the first condition as POI information pairs. The POI information pair acquisition unit 12 stores the acquired information of the POI information pair in the POI information pair management table. FIG. 3 is a diagram showing an example of the POI information pair management table. The POI information pair stored in the POI information pair management table may be presented to an operator, for example. In this case, the operator may visually check, for example, the POI information pair presented on the display or the like, and delete the POI information pair determined to be unlikely to indicate the same POI from the POI information pair management table. By using the visual check by the operator in combination, the accuracy improvement of the POI information pair stored in the POI information pair management table can be expected.

  The category pair acquisition unit 13 acquires a category pair based on a combination of category information included in each of the POI information included in the POI information pair acquired by the POI information pair acquisition unit 12. In the present embodiment, the category pair acquisition unit 13 determines the category based on the number of appearances of each category information and the number of appearances of each combination of category information in the plurality of pairs of POI information acquired by the POI information pair acquisition unit 12. Get a pair.

  First, the category pair acquisition unit 13 refers to the plurality of POI information pairs acquired by the POI information pair acquisition unit 12 (each POI information pair stored in the POI information pair management table of FIG. 3) to make a category pair. Extract candidates. Specifically, the category pair acquisition unit 13 refers to the POI information pair in the first line of the POI information pair management table of FIG. 3 to pair the category information “sushi store” with the category information “sushi”. Extract as a category pair candidate. Similarly, the category pair acquisition unit 13 sets a pair of category information “commercial facility” and category information “shopping mall” as a category pair candidate from the POI information pair in the second and third rows of the POI information pair management table. Extract. The category pair acquisition unit 13 stores the information on the category pair candidates thus extracted in the category pair candidate management table.

  FIG. 4 is a diagram showing an example of a category pair candidate management table. The category pair candidate management table stores one category information X and the other category information Y. The category pair candidate management table also stores, for each category pair candidate, the number of appearances of category information X, the number of appearances of category information Y, the number of simultaneous appearances, and the Dice coefficient calculated by processing of the category pair acquisition unit 13 described later. Be done.

  The category pair acquisition unit 13 counts the number of appearances for each category information by referring to each POI information pair stored in the POI information pair management table, and counts the counted value as “category information X in the category pair candidate management table. It is stored in the appearance frequency column and the category information Y appearance frequency column. In addition, the category pair acquisition unit 13 counts the number of appearances for each combination of category information, and stores the counted value in the “number of simultaneous appearances” column of the category pair candidate management table. For example, the simultaneous appearance frequency "280" of category pair candidates in the first line of the category candidate management table in FIG. 4 indicates the number of POI information pairs having a combination of category information "sushi store" and category information "sushi". .

Subsequently, for each category pair candidate, the category pair acquisition unit 13 simultaneously appears the number of appearances of one category information X (number of appearances of category information X) and the number of appearances of the other category information Y (number of appearances of category information Y). Based on the number of times, the Dice coefficient is calculated by the following (Equation 3). Here, | X | represents the number of appearances of category information X, | Y | represents the number of appearances of category information Y, and | X∩Y | represents the number of simultaneous appearances.
(Equation 3): Dice coefficient = 2 × | X∩Y | / (| X | + | Y |)

  The Dice coefficient calculated by the above (Equation 3) indicates the similarity between one category information and the other category information. Specifically, the Dice coefficient is considered to be larger as the degree of similarity between category information is larger. That is, as the Dice coefficient is larger, there is a relation that two category information is more likely to indicate the same category. Therefore, the category pair acquisition unit 13 may acquire the category pair based on the comparison between the Dice coefficient and a predetermined threshold value, using the property of such Dice coefficient. For example, even if the category pair acquisition unit 13 acquires a pair of one category information and the other category information as a category pair when the Dice coefficient is equal to or greater than a predetermined threshold (for example, 0.7 or the like). Good.

  In this case, in the example of FIG. 4, a pair of the category information "sushi store" having the Dice coefficient "0.903" and the category information "sushi" and the category information "commercial" having the Dice coefficient "0.778" A pair of “facility” and category information “shopping mall” is acquired as a category pair. On the other hand, a pair of the category information “Japanese-style pub” having the Dice coefficient of “0.050” and the category information “restaurant” is not acquired as a category pair.

  The category pair acquisition unit 13 generates a category pair management table (not shown) that stores only category pairs by deleting from the category pair candidate management table information of category pair candidates whose Dice coefficient is less than the threshold value. Good. However, as illustrated in FIG. 4, the category pair acquisition unit 13 may end the process when the Dice coefficient of each category pair candidate is stored in the “Dice coefficient” column of the category pair candidate management table. Also in this case, by referring to the “Dice coefficient” column of the category pair candidate management table, it becomes possible to specify a category pair (that is, a pair whose Dice coefficient is equal to or more than the threshold). That is, the category pair acquisition unit 13 substantially acquires the category pair.

  The category pair acquisition unit 13 may omit the above-described determination process using the Dice coefficient. In this case, the category pair acquisition unit 13 may immediately acquire category information of each piece of POI information included in the POI information pair as a category pair. However, by performing the above-described determination process using the Dice coefficient, it can be expected to extract category pairs with higher accuracy.

In the above example, the category pair acquisition unit 13 uses the Dice coefficient as an index indicating the similarity between one category information and the other category information, but the category pair acquisition unit 13 uses an index other than the Dice coefficient. It is also good. For example, the category pair acquisition unit 13 may use the Jaccard coefficient, the Simpson coefficient, and the like calculated by the following (Expression 4) and (Expression 5) instead of the Dice coefficient. Here, | X∪Y | represents the number of category pair candidates including at least one of the category information X and the category information Y among the category pair candidates stored in the category pair candidate management table. min (x, y) represents the smaller value of x and y.
(Equation 4): Jaccard coefficient = | X∩Y | / | X∪Y |
(Equation 5): Simpson coefficient = | X∩Y | / min (| X |, | Y |)

  By the processing of the POI information pair acquiring unit 12 and the category pair acquiring unit 13 described above, category pairs indicating the same category are acquired (extracted).

  The comparison target acquisition unit 14 acquires first POI information (first location information) and second POI information (second location information) to be compared. The first POI information and the second POI information are input by an operator, for example.

  Further, as described above, the first condition used by the POI information pair acquisition unit 12 is set to a relatively severe condition. Therefore, there is a possibility that POI information pair candidates indicating the same POI are not actually extracted as POI information pairs by the POI information pair acquisition unit 12. Therefore, the comparison target acquiring unit 14 selects, from among the POI information pair candidates included in the POI information pair candidate management table, the POI information pair candidate not acquired as a POI information pair by the POI information pair acquiring unit 12 as the first POI information and You may acquire as 2nd POI information.

  The comparison target acquiring unit 14 pairs the first POI information (name A, category information A, latitude A, longitude A) and second POI information (name B, category information B, latitude B, longitude B) thus acquired. Information of (hereinafter, "compared pair") is stored in the compared pair management table.

  FIG. 5 is a diagram showing an example of the comparison target pair management table. The comparison target pair management table stores information of the distance (dist) and the longest common subsequence ratio (lcsr) calculated by the POI information pair acquiring unit 12 as information of the comparison target pair, together with the first POI information and the second POI information. doing. The comparison target pairs in the first to third rows of the comparison target pair management table of FIG. 5 are POI information pair candidates that were not acquired as POI information pairs by the POI information pair acquisition unit 12 (POI information pair candidate management table of FIG. This corresponds to the POI information pair candidate in the first, second, and fourth lines.

  The determination unit 15 determines the first POI information and the second POI information based on the category information A included in the first POI information, the category information B included in the second POI information, and the category pair acquired by the category pair acquisition unit 13. Determine whether or not the same POI is indicated.

  First, the determination unit 15 determines whether the pair of the category information A and the category information B corresponds to the category pair acquired by the category pair acquisition unit 13. When the category pair management table described above is generated, the determination unit 15 refers to the category pair management table to determine whether the pair of the category information A and the category information B corresponds to the category pair. Can be determined. On the other hand, when the category pair management table is not generated and the Dice coefficient calculated by the category pair acquisition unit 13 is stored in the “Dice coefficient” column of the category pair candidate management table, the determination unit 15 Whether or not a pair of category information A and category information B corresponds to a category pair can be determined as follows. That is, if the pair of category information A and category information B is stored in the category pair candidate management table and the Dice coefficient of the pair is equal to or greater than the threshold, determination unit 15 determines category information A and category information. It can be determined that the pair with B corresponds to the category pair. If the category information A and the category information B are the same character string, the determination unit 15 immediately determines that the pair of the category information A and the category information B corresponds to the category pair. It is also good.

  When determining that the pair of the category information A and the category information B corresponds to a category pair, the determination unit 15 assigns a category flag “1” to the comparison target pair. On the other hand, determination unit 15 assigns category flag “0” to the comparison target pair to which category flag “1” is not assigned.

  Next, the determination unit 15 determines that the names are not identical (i.e., lcsr <1) between the comparison target pairs (the first POI information and the second POI information), and one of the names includes the other's name. It is determined whether or not the predetermined character position of includes a predetermined character. For example, the determination unit 15 determines whether or not a predetermined character string “store” is included as the final character string of one of the names. If “store” is included in the final character string of one of the names, the POI having one of the names may be a tenant within the POI having the other name (for example, a commercial facility or the like). As described above, by comparing the names, the POI having one name and the POI having the other name have a special relationship (in the above example, one is a tenant such as an individual store, and the other accommodates a tenant). In some cases, it can be inferred to have a relationship that is an owner facility. Therefore, the determination unit 15 makes it possible to prevent the comparison target pair whose special relationship is inferred based on the comparison of such names from being erroneously determined to indicate the same POI. The exception flag "1" is given. On the other hand, the determination unit 15 assigns the exception flag "0" to the comparison target pair to which the exception flag "1" is not assigned. Note that the above determination is an example, and the determination unit 15 determines in the same manner as described above including a character string other than "store" described above as a predetermined character string (for example, "branch", "branch", etc.) You may

  When the number of characters of the name A of the first POI information and the number B of the name B of the second POI information is the same, the determination unit 15 determines the name A as one name and the name B as one name. Both of the case determination and the determination may be performed. Moreover, the determination part 15 may perform the pre-processing with respect to the name A and the name B, when performing the said determination. The pre-processing is, for example, processing for disambiguating in order to appropriately compare character strings, and may include deletion of space notation included in the character strings, known stemming processing, and the like.

  FIG. 6 is a diagram showing a state in which a category flag and an exception flag are added to each comparison target pair stored in the comparison target pair management table by the process of the determination unit 15 described above. In FIG. 6, illustration of position information (latitude and longitude) of each POI information is omitted.

The determination unit 15 determines whether or not the comparison target pair indicates the same POI, based on the distance (dist) between POIs of the comparison target pair, the longest common partial sequence ratio (lcsr), the category flag, and the exception flag. Determine The determination unit 15 may determine whether the comparison target pair indicates the same POI based on a predetermined rule. For example, the determination unit 15 may set a POI to be compared using the second condition (condition for POI name (lcsr ≧ 0.8) and condition for POI position (dist ≦ 100)) and the flag condition. It may be determined that it is an information pair.
(Second condition)
・ Lcsr ≧ 0.8
・ Dist ≦ 100
(Flag condition)
-Category flag = 1
・ Exception flag = 0

  Here, the second condition is a condition relaxed than the first condition used for the determination of the POI information pair acquisition unit 12. As described above, the determination unit 15 relaxes the determination condition (second condition) on the POI name and position as compared with the determination condition (first condition) used by the POI information pair acquisition unit 12 to obtain the POI information pair. Occurrence of the extraction leakage of On the other hand, the determination unit 15 can suppress the occurrence of the erroneous extraction of the POI information pair caused by the relaxation of the determination condition on the name and position of the POI by performing the determination in combination with the above-described flag condition. By performing the determination based on the identity of the category in this manner, extraction omission can be suppressed while suppressing erroneous extraction of the POI information pair, and the POI information pair can be appropriately extracted.

  Further, the determination unit 15 may perform the above determination using a determination model generated in advance by known machine learning. Such a determination model is generated, for example, by performing machine learning using as a learning data a pair of POI information whose whether or not to indicate the same POI is known. The judgment model inputs, for example, distance (dist), longest common subsequence ratio (lcsr), category flag, and exception flag as explanatory variables, and whether or not the same POI is indicated as an objective variable (for example, the same POI) This is a model that outputs "1" when shown, and a variable that takes "0" when not showing the same POI.

  Next, the operation of the server 10 will be described with reference to the flowchart shown in FIG. The processes of steps S1 and S2 are processes for acquiring category pairs. The processes of steps S3 and S4 are processes for executing name identification determination of the first POI information and the second POI information to be compared using the category pair acquired in step S2.

  In step S1, the POI information pair acquisition unit 12 acquires a POI information pair indicating the same POI from among a plurality of POI information. In the present embodiment, the POI information pair acquiring unit 12 determines each POI information pair candidate stored in the POI information pair candidate management table (see FIG. 2) by the determination using the first condition on the POI name and position. It is determined whether or not the same POI is indicated.

  In step S2, the category pair acquisition unit 13 acquires a category pair which is a pair of category information indicating the same category. In the present embodiment, the category pair acquisition unit 13 acquires category pair candidates based on the POI information pair acquired by the POI information pair acquisition unit 12. Then, for each category pair candidate, the category pair acquisition unit 13 displays the number of appearances of one category information (number of appearances of category information X), the number of appearances of category information of the other (number of appearances of category information Y) and the number of simultaneous appearances. Get category pair based on. More specifically, for example, if the Dice coefficient calculated by the above (Equation 3) is equal to or greater than a predetermined threshold (for example, 0.7 or the like), the category pair acquisition unit 13 The pair with the other category information is acquired as a category pair.

  In step S3, the comparison target acquisition unit 14 acquires the first POI information and the second POI information to be compared. For example, among the POI information pair candidates included in the POI information pair candidate management table, the comparison target acquiring unit 14 selects the POI information pair candidate not acquired as a POI information pair by the POI information pair acquiring unit 12 as the first POI information and Acquire as second POI information.

  In step S4, based on the category information included in the first POI information, the category information included in the second POI information, and the category pair acquired by the category pair acquisition unit 13, the determination unit 15 determines the first POI information and the first POI information. 2) It is determined whether or not the POI information indicates the same POI. For example, when the pair of the first POI information and the second POI information (the comparison target pair) satisfies the above-described second condition and flag condition, the determining unit 15 indicates that the first POI information and the second POI information indicate the same POI. judge.

  In the server 10 described above, the POI information pair indicating the same POI is acquired from the plurality of POI information. And based on the combination of the category information contained in the said POI information pair, the category pair which is a pair of the category information which shows the same category is acquired. By linking the identity of POI and the identity of categories in this manner, category pairs indicating the same category can be generated with high accuracy. Such category pairs can be used to accurately determine the identity of POI information. In the present embodiment, category pairs may be used for determination based on the category flag described above. Therefore, according to the server 10, it is possible to more accurately determine the identity of the POI information.

  Further, the category pair acquisition unit 13 acquires a category pair based on the appearance frequency for each category information and the appearance frequency for each combination of category information in the POI information pair acquired by the POI information pair acquisition unit 12. In the present embodiment, as an example, the category pair acquisition unit 13 acquires (extracts) a category pair based on the comparison between the Dice coefficient calculated by the above (Equation 3) and the threshold. With this configuration, category pairs can be extracted more accurately.

  In addition, the server 10 acquires the first POI information and the second POI information to be compared, the comparison target acquiring unit 14, the category information included in the first POI information, the category information included in the second POI information, and the category pair And a determination unit that determines whether the first POI information and the second POI information indicate the same POI based on the category pair acquired by the acquisition unit. According to such a configuration, the identity of the category of the first POI information and the category of the second POI information is determined based on the category pair acquired by the category pair acquisition unit 13, and the POI information is determined on the basis of the determination result. The identity can be determined appropriately.

  Specifically, in the present embodiment, when the combination of the category information of the pair of POI information to be compared matches the category pair, for example, “sushi store” and “sushi”, the POI information The category flag "1" is set to the pair. On the other hand, when the combination of the category information of the pair of POI information does not match the category pair as in, for example, “sushi store” and “shopping mall”, the category flag “0” is set to the pair of POI information. . Thereafter, as described above, the identity of the POI information is determined based on the category flag as well. As described above, the determination based on the identity of the categories makes it possible to accurately determine the identity of the POI information.

  Further, the POI information pair acquiring unit 12 determines whether or not a first predetermined condition regarding the name and position is satisfied among the plurality of POI information, and the POI information pair is determined based on the result of the determination. get. The determination unit 15 determines whether or not a predetermined second condition for the name and position is satisfied between the first POI information and the second POI information, and the first POI information is also based on the result of the determination. It is determined whether the second POI information and the second POI information indicate the same POI. Here, the second condition is a condition relaxed than the first condition. According to this configuration, as described above, it is possible to suppress the occurrence of the extraction omission of the POI information pair and to suppress the occurrence of the false extraction of the POI information pair due to the relaxation of the determination condition for the name and the position.

  In addition, the determination unit 15 does not have the same name between the first POI information and the second POI information, and one name includes the other name, and is determined in advance at a predetermined character position of one name. It is determined whether the designated character is included, and it is determined whether the first POI information and the second POI information indicate the same POI based on the result of the determination. In the present embodiment, the determination unit 15 performs such determination by setting the above-described exception flag. According to this configuration, it is possible to appropriately prevent the erroneous extraction of the POI information pair when it is presumed that the first POI information and the second POI information are in a special relationship that does not indicate the same POI.

  As mentioned above, although one embodiment of the present invention was described, the present invention is not limited to the above-mentioned embodiment. For example, the server 10 may not include the comparison target acquisition unit 14 and the determination unit 15. In this case, even if the category pair generated by the server 10 is output to another device, and the other device executes the processing of the comparison target acquisition unit 14 and the determination unit 15 described above using the category pair. Good.

  The block diagram used for the explanation of the above-mentioned embodiment has shown the block of a functional unit. These functional blocks (components) are realized by any combination of hardware and / or software. Moreover, the implementation means of each functional block is not particularly limited. That is, each functional block may be realized by one physically and / or logically coupled device, or directly and / or indirectly two or more physically and / or logically separated devices. It may be connected (for example, wired and / or wirelessly) and realized by the plurality of devices.

  For example, the server 10 in the above embodiment may function as a computer that performs the processing of the server 10 in the above embodiment. FIG. 8 is a diagram showing an example of the hardware configuration of the server 10 according to the present embodiment. The above-described server 10 may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007 and the like.

  In the following description, the term "device" may be read as a circuit, a device, a unit, or the like. The hardware configuration of the server 10 may be configured to include one or more of the devices illustrated in FIG. 8 or may be configured without including some devices.

  Each function in the server 10 performs a calculation by causing the processor 1001 to read predetermined software (program) on hardware such as the processor 1001, the memory 1002, etc., communication by the communication device 1004, data in the memory 1002 and the storage 1003 This is realized by controlling the reading and / or writing of

  The processor 1001 operates, for example, an operating system to control the entire computer. The processor 1001 may be configured by a central processing unit (CPU) including an interface with a peripheral device, a control device, an arithmetic device, a register, and the like.

  Also, the processor 1001 reads a program (program code), a software module, and / or data from the storage 1003 and / or the communication device 1004 to the memory 1002, and executes various processing according to these. As a program, a program that causes a computer to execute at least a part of the operations described in the above embodiments is used. For example, the determination unit 15 of the server 10 may be realized by a control program stored in the memory 1002 and operated by the processor 1001, or may be realized similarly for other functional blocks shown in FIG. The various processes described above have been described to be executed by one processor 1001, but may be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The program may be transmitted from the network via a telecommunication line.

  The memory 1002 is a computer readable recording medium, and includes, for example, at least one of a ROM (Read Only Memory), an EPROM (Erasable Programmable ROM), an EEPROM (Electrically Erasable Programmable ROM), and a RAM (Random Access Memory). It may be done. The memory 1002 may be called a register, a cache, a main memory (main storage device) or the like. The memory 1002 can store a program (program code), a software module, and the like that can be executed to implement the information processing method (the procedure and the like shown in the flowchart of FIG. 7) according to the above embodiment.

  The storage 1003 is a computer readable recording medium, and is, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (for example, a compact disk, a digital versatile disk, Blu-ray A (registered trademark) disk, a smart card, a flash memory (for example, a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, and the like may be used. The storage 1003 may be called an auxiliary storage device. The above-described storage medium may be, for example, a database including the memory 1002 and / or the storage 1003, a server, or any other suitable medium.

  The communication device 1004 is hardware (transmission / reception device) for performing communication between computers via a wired and / or wireless network, and is also called, for example, a network device, a network controller, a network card, a communication module, or the like.

  The input device 1005 is an input device (for example, a keyboard, a mouse, a microphone, a switch, a button, a sensor, and the like) that receives external input. The output device 1006 is an output device (for example, a display, a speaker, an LED lamp, etc.) that performs output to the outside. The input device 1005 and the output device 1006 may be integrated (for example, a touch panel).

  Also, each device such as the processor 1001 and the memory 1002 is connected by a bus 1007 for communicating information. The bus 1007 may be configured by a single bus or may be configured by different buses among the devices.

  The server 10 is configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), and a field programmable gate array (FPGA). Some or all of the functional blocks may be realized by the hardware. For example, processor 1001 may be implemented in at least one of these hardware.

  Although the present invention has been described above in detail, it is apparent to those skilled in the art that the present invention is not limited to the embodiments described herein. The present invention can be implemented as a modified and changed embodiment without departing from the spirit and scope of the present invention defined by the description of the claims. Accordingly, the description in the present specification is for the purpose of illustration and does not have any limiting meaning on the present invention.

  As long as there is no contradiction, the processing procedure of each aspect / embodiment, flow chart, etc. which were explained to this specification may be replaced. For example, for the methods described herein, elements of the various steps are presented in an exemplary order and are not limited to the particular order presented.

  The input / output information may be stored in a specific place (for example, a memory) or may be managed by a management table. Information to be input or output may be overwritten, updated or added. The output information etc. may be deleted. The input information or the like may be transmitted to another device.

  The determination may be performed by a value (0 or 1) represented by one bit, may be performed by a true / false value (Boolean: true or false), or may be compared with a numerical value (for example, a predetermined value). Comparison with the value).

  Each aspect / embodiment described in the present specification may be used alone, may be used in combination, and may be switched and used along with execution. In addition, notification of predetermined information (for example, notification of "it is X") is not limited to what is explicitly performed, but may be performed implicitly (for example, not notifying of the predetermined information) Good.

  Software may be called software, firmware, middleware, microcode, hardware description language, or any other name, and may be instructions, instruction sets, codes, code segments, program codes, programs, subprograms, software modules. Should be interpreted broadly to mean applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, functions, etc.

  Also, software, instructions and the like may be transmitted and received via a transmission medium. For example, software may use a wireline technology such as coaxial cable, fiber optic cable, twisted pair and digital subscriber line (DSL) and / or a website, server or other using wireless technology such as infrared, radio and microwave When transmitted from a remote source, these wired and / or wireless technologies are included within the definition of transmission medium.

  The information, signals, etc. described herein may be represented using any of a variety of different techniques. For example, data, instructions, commands, information, signals, bits, symbols, chips etc that may be mentioned throughout the above description may be voltage, current, electromagnetic waves, magnetic fields or particles, light fields or photons, or any of these May be represented by a combination of

  The terms described in the present specification and / or the terms necessary for the understanding of the present specification may be replaced with terms having the same or similar meanings.

  The terms "system" and "network" as used herein are used interchangeably.

  In addition, the information, parameters, and the like described in the present specification may be represented by an absolute value, may be represented by a relative value from a predetermined value, or may be represented by corresponding other information. .

  The names used for the parameters described above are in no way limiting. In addition, the formulas etc. that use these parameters may differ from those explicitly disclosed herein.

  As used herein, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

  Any reference to an element using the designation "first," "second," etc. as used herein does not generally limit the quantity or order of those elements. These designations may be used herein as a convenient way of distinguishing between two or more elements. Thus, reference to the first and second elements does not mean that only two elements can be taken there, or that in any way the first element must precede the second element.

  Insofar as "including", "including", and variations thereof are used in the present specification or claims, these terms as well as the term "comprising" are inclusive. Intended to be Further, it is intended that the term "or" as used in the present specification or in the claims is not an exclusive OR.

  In the present specification, a plurality of devices are also included unless a context or technically apparent device is shown as having only one.

  Throughout this disclosure, unless the context clearly indicates otherwise, it is intended to include the plural.

  DESCRIPTION OF SYMBOLS 10 ... Server, 11 ... Storage part, 12 ... POI information pair acquisition part, 13 ... Category pair acquisition part, 14 ... Comparison object acquisition part, 15 ... Determination part, 1001 ... Processor, 1002 ... Memory, 1003 ... Storage, 1004 ... Communication device, 1005 ... input device, 1006 ... output device.

Claims (5)

A location information pair acquisition unit for acquiring a pair of location information indicating the same location from among a plurality of location information each including category information indicating a category of location;
A category pair acquisition unit for acquiring a category pair which is a pair of category information indicating the same category based on a combination of category information included in each of the location information included in the pair acquired by the location information pair acquisition unit; ,
A name identification support device provided with The category pair acquisition unit acquires the category pair based on the number of appearances for each category information and the number of appearances for each combination of category information in the location information pairs acquired by the location information pair acquisition unit.
The name identification support device according to claim 1. A comparison target acquisition unit that acquires first place information and second place information to be compared;
The first place information and the second place are based on the category information included in the first place information, the category information included in the second place information, and the category pair acquired by the category pair acquisition unit. A determination unit that determines whether the information indicates the same place;
Further comprising
The name identification support device according to claim 1. The place information further includes information indicating the name of the place and information indicating the position of the place,
The place information pair acquiring unit determines whether a first predetermined condition regarding a name and a position is satisfied among the plurality of place information, and the pair of the place information is based on a result of the determination. Get
The determination unit determines whether or not a second predetermined condition regarding a name and a position is satisfied between the first place information and the second place information, and is also based on the result of the determination. Determining whether the first place information and the second place information indicate the same place,
The second condition is a condition relaxed than the first condition,
The name identification support device according to claim 3. The determination unit does not have the same name between the first place information and the second place information, and one of the names includes the other's name, and the predetermined character position of one of the names is determined. It is determined whether or not a predetermined character is included, and it is determined whether the first place information and the second place information indicate the same place based on the result of the determination.
The name identification support device according to claim 3 or 4.

Images