JP2000276472A - Method and device for similar information collation and recording medium for recording similar information collation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a similar information collation device which discriminates the similarity of information by pattern collation in the case that a noise is generated in a minute pattern partially coinciding in a collation pattern to break up the minute pattern. SOLUTION: A similar information collation device 1 is provided with a means 10 which generates first and second patterns represented by the positions and features of the elements from information to be collated, a means 20 which generates a collation map 30 consisting of collation positions whose coordinates are pairs of positions of first and second elements having the same features in the first and second patterns, a means 40 which evaluates the continuity of each of routes where neighboring collation positions in the collation map 30 are successively connected, and a means 40 which discriminates the degree of coincidence between the first and second patterns on the basis of the continuity of each route.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a similar information collating apparatus and a similar information collating method for determining the similarity of information that can be expressed as a pattern, and more particularly to a technique for collating various patterns processed by a computer.

[0002]

2. Description of the Related Art Information such as characters, figures, images, sounds, or general symbols can be represented by the position and characteristics of the elements constituting the information. As described above, information in which the features of each element are arranged in a spatiotemporal manner when focusing on the features of the elements is called a so-called pattern. Conventionally, various pattern information processing methods have been proposed for processing this pattern by a computer. In the process of pattern information processing, it is frequently required to search for a similar character string in a situation where the degree of matching of similar information should be evaluated, for example, in the case of character information. Therefore, as a method of determining the similarity of information, a pattern matching method is known in which elements of information are expressed as patterns and the relevance between the patterns is evaluated.

As an example of a conventional pattern matching method, a character string matching system will be described. For example, a conventional high-speed full-text search technology using a character string matching system is described in “2nd.
Elemental Technology for High-speed Full-text Search in Nippon Express, The Index Processing That Holds the Key, "Nikkei Byte, October 1996, page 158
-167. The conventional typical collation system described in this reference divides the collation character string and the collated character string into small portions of fixed length. here,
The usage of the terms “collation character string” and “character string to be collated” will be briefly described. For example, when a character string B similar to the character string A is found in the document C, the character string A becomes “collation character string”. , And the character string B in the document C is a “character string to be verified”.
It is. Next, the collation system determines whether the minute part of the collation character string is included in the minute part character string group of the collated character string, and determines that the collated character string including the minute part character string is similar to the collation character string. Is output as a converted character string.

A collation system of this type determines whether there is a character string in which a minute part in the character string completely matches. Therefore, when a part of a character string is missing, a part of a character string is replaced with another character string, or a case where another character string is mixed in a character string, Alternatively, when a local deformation occurs in the collated character string, it is determined that the character strings are not collated because the minute character strings around the deformed portion do not match. As described above, the first type of collation system of the related art has a disadvantage that local deformation of a character string cannot be tolerated.

[0005]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems of the conventional collating system, and has been described in the context of a similar information collating apparatus which determines the similarity of information. In the second pattern, even when a minute pattern that partially matches is partially dispersed, replaced with another pattern, or mixed in with another pattern, the pattern matching is performed even if it is dispersed throughout the pattern. It is an object of the present invention to provide a similar information collation device, a similar information collation method, and a recording medium on which a similar information collation program is recorded, which can determine the similarity of information by performing the same.

[0006]

In order to achieve the above object, the present invention introduces a concept of continuity which tracks a pattern matching position and allows a separated matching position, and evaluates this continuity. To prevent omission of verification. FIG. 1 is a diagram showing the principle of the present invention. The similarity information matching device 1 for determining the similarity of information according to the present invention uses the first information and the second information to be compared as a pattern represented by the position and characteristic of an element of the information. Pattern generating means 10 for generating a first pattern corresponding to information and a second pattern corresponding to the second information; and a second pattern having the same characteristics among the first pattern and the second pattern. A collation map generating means 20 for producing a collation map 30 composed of a collation position using a pair of respective positions of a first element belonging to one pattern and a second element belonging to the second pattern as coordinates, A continuity evaluating means for evaluating the continuity of the route for each of the routes in which the matching positions in the vicinity in the matching map are sequentially connected, and the continuity evaluating means for evaluating the continuity evaluated for each of the routes. Patterns and the determining pattern matching means for matching of the second pattern 50
And

[0007] The collation map creating means 20 is characterized in that the collation positions are individually created for a plurality of combinations of the first element and the second element having the same characteristics. Further, the pattern matching means 50 sets, for each of the matching positions, the highest continuity among the continuities evaluated for the path passing through the matching position as an evaluation value of the matching position, Means for calculating a degree of coincidence between the first pattern and the second pattern based on an evaluation value set for each collation position.

[0008] Further, the pattern generating means 10 is a synonymous element having a feature capable of replacing a feature of the at least part of the original element with respect to at least a part of the information represented as the pattern. Means for generating
Means for generating the pattern so that the synonymous elements are listed at the same time as the original element. The matching map generating means 20, the continuity evaluating means 30, and the pattern matching means 40 , Characterized in that the synonymous elements listed at the same time are adapted to be processed in parallel with the original elements.

The collation map generation means 20 is configured to include a means for judging that the element has the same characteristic when the element has a characteristic expressing a numerical value and the value represented by the numerical value matches. Is also good. FIG. 2 shows an operation of a similarity information matching method for determining the similarity of information by comparing a first pattern and a second pattern represented by the positions and characteristics of information elements to achieve the object of the present invention. It is a flowchart. As shown in the drawing, the similar information collating method of the present invention comprises the steps of inputting the first pattern and the second pattern (step 1), and includes the steps of inputting the first pattern and the second pattern. Detecting a first element belonging to the first pattern and a second element belonging to the second pattern having the same characteristics (step 2)
A matching map generating step of creating a matching map having coordinates of pairs of the positions of the detected first element and the detected second element (step 3); and a step of generating the matching map in the vicinity of the matching map. A route generation step of generating a route by sequentially connecting positions (step 4), a continuity evaluation step of evaluating the continuity of the route for each of the generated routes (step 5), A pattern matching step (step 6) of determining a matching degree between the first pattern and the second pattern based on the evaluated continuity.

Further, in the similar information collating system for judging the similarity of information, the similar information collating apparatus and method of the present invention for judging the similarity of information according to the present invention provide a computer-readable storage medium storing a program ( Software). Therefore, the present invention
It includes a computer-readable recording medium that records a similarity information collation program that determines the similarity of information. The similarity information collation program extracts a first pattern corresponding to the first information from the first information and the second information to be collated as a pattern represented by a position and a characteristic of an element of the information. A pattern generation code for generating a second pattern corresponding to the second information; a first element belonging to the first pattern having the same feature in the first pattern and the second pattern; Second
The collation map generation code for creating a collation map composed of the collation positions each having a pair of the positions of the second elements belonging to the pattern and the collation positions adjacent in the collation map are sequentially connected. Pattern matching for evaluating the continuity of the route for each of the determined routes, and pattern matching for determining the degree of matching between the first pattern and the second pattern based on the continuity evaluated for each of the routes And a code.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A character string collating system according to one embodiment of the present invention will be described below with reference to the accompanying drawings. The character string matching system according to the present embodiment is a system for presenting a searched document including a sentence similar to the search sentence input by the operator from the searched documents stored in the searched document file to the operator.

FIG. 3 is a schematic configuration diagram of a character string collating system according to an embodiment of the present invention, and FIG. 4 is an operation flowchart of the character string collating system. The character string collation system includes a collation data generator 110 that receives a search sentence input by an operator in step 10. The collation data generation unit 110 converts the search sentence into a collation character string as collation data suitable for collation (step 20). In step 20, the character string matching system extracts the searched document from the searched document file 140, and includes the matched character string suitable for matching with the matched character string and the document identification number of the searched document to which the matched character string belongs. It further has a data-to-be-collated generating unit 130 that generates data to be collated.

The collation character string as collation data and the collation character string as collation data are patterns expressing character information, among patterns expressing various information. Each character in the string corresponds to an element of the pattern.
The element is represented by a character code as a characteristic of the character and a position of the character in the character string. Also,
The character string collation system further includes a collation map generation unit 150, and the collation map generation unit 150
Receiving the collation character string from 0 and the collation character string from the collation data generation unit 130, a common character is detected (step 30), and a collation map is created and output (step 4).
0).

In the collation map, the positions of the common characters included in the collation character string and the collated character string in the collation character string and the collated character string are represented as X coordinates and Y coordinates, respectively (X, Y). Y) is a map having Y) as a collation position.
The collation map can be constructed as a two-dimensional map when a character string is adopted as a pattern. The character string collation system further includes a continuity evaluation unit 160 and a search result output unit 170. The continuity evaluation unit 160 receives the collation map created and output by the collation map generation unit 150, collates the collation character string with the collated character string, and passes the collation result to the search result output unit. Therefore, the continuity evaluation unit 1
60 sequentially forms a path including a series of collation positions by sequentially tracking collation positions near the collation position in the collation map (step 50), and calculates a continuity value for each path (step 60). ), As an evaluation value for each collation position, the highest continuity value among the continuity values of the route passing through the collation position is selected (step 70), and the collation position of each character of the collation character string is determined. The evaluation values are totaled (step 80). This total result indicates the similarity between the collation character string and the collated character string.

The search result output unit 170 is a continuity evaluation unit 1
Receiving the collation result from 60, information relating to the retrieved document including the collated character string determined to be similar to the collated character string is extracted from the retrieved document file 140 and notified to the operator (step 90). Hereinafter, a character string collation system according to an embodiment of the present invention will be described in detail.

FIG. 5 is a block diagram of the collation data generation unit 110 of the character string collation system in this embodiment. As shown in the figure, a collation data generation unit 110 inputs a search sentence, and outputs a search sentence expansion unit 111 by referring to a search sentence expansion dictionary 120. A search sentence normalizing unit 112 for converting into an extended search sentence, and a numeric expression replacing unit for inputting the normalized extended search sentence, converting the numerical expression part into the same format, and outputting a collation character string as final collation data 113.

The search sentence expansion unit 111 receives a search sentence (eg, “speed-up of document search”) from the operator and generates a character string that can replace a character string (eg, “search”) in the search sentence. It is determined whether it exists in the search sentence expansion dictionary 120, and a replaceable character string (for example, “extract”)
Is present in the search sentence expansion dictionary 120, the replaceable character string (“extraction”) is converted into a character string (“search”).
Is added to the search sentence as a synonym for, and an extended search sentence is output. If there is no replaceable character string, the input search sentence is output as it is as an extended search sentence.
The extended search sentence in this case is expressed as “speedup of document {search | extraction}”, where {character string a | character string b} is an extended part, and character string a and character string b represents a synonym.

The search sentence normalizing unit 112 includes a search sentence expanding unit 1
The extended search sentence output from 11 is input, characters in the extended search sentence are normalized, and a normalized extended search sentence is output.
Character normalization is, for example, conversion of half-width characters of alphanumeric kana to full-width characters, conversion of lowercase letters to uppercase letters,
Alternatively, it means processing such as deletion of noise characters that should be ignored when searching for punctuation marks, line feed control characters, extended sounds, and the like.

The numerical expression replacement unit 113 receives the normalized expanded search sentence output from the search sentence normalizing unit 112, and determines whether or not a partial character string quantitatively expressed by a numerical value exists in the normalized expanded search sentence. I do. If there is a substring expressed quantitatively, convert the substring to a value,
Creates and outputs a collation character string as final collation data in which the quantitative expression part in the normalized extended search sentence is replaced by a value.

FIG. 6 is a configuration diagram of the collated data generation unit 130 of the character string collation system in this example. As shown in the figure, the matching data generation unit 130 has a searched document reading unit 131 that reads all the searched documents required for the search from the searched document file 140 and outputs the read documents. The data-to-be-verified unit 130 is a unit for normalizing the document to be searched 132
And a numerical expression replacement unit 133.

The search target document normalizing section 132 receives the search target document from the search target document reading section 131, normalizes characters in the search target document, and outputs a normalized search target document. The normalization of characters is as described in the search sentence normalization unit 112. In addition, the numerical expression replacement unit 133
Receives the normalized search target document from the search target document normalization unit 132, and if there is a substring quantitatively represented by a numerical value in the normalized search target document, converts the substring into a value; It creates and outputs a collated character string as final collated data in which the quantitative expression part in the normalized searched document is replaced by a value.

Next, the function of the matching map generator 150 of the character string matching system according to one embodiment of the present invention will be described in detail. As shown in FIG.
Are the collation data generation unit 110 and the collation target data generation unit 1
And a function of receiving a collation character string as collation data and a collation character string as collation data from each, and generating a collation map.

First, the collation map generation unit 150 first generates a character included in both the collation character string and the collated character string,
That is, a common character is detected. For example, if the collation character string is “speed up document search” and the collated character string is “search for high-speed document”, the common characters are “high”,
"Quick", "sentence", "call", "no", "inspection" and "search". Next, a collation map is generated that includes collation positions where the position of the common character in the collation character string is the Y coordinate and the position of the common character in the collation character string is the X coordinate.

FIG. 7 is an explanatory diagram showing, by way of example, a collation map generated for the collation character string and the collated character string, so that the concept of the collation map is better understood. . In the figure, a point indicated by “○” corresponds to the collation position. In this example, for the sake of simplicity, it is assumed that the collated character string does not include a synonym. On the other hand, as described above, in the search sentence expansion unit 111, the collation character string “speed up document search” is expanded to include a synonym in the form of the expanded search sentence “speed up document {search | extraction}”. , A matching map as shown in FIG. 8 is obtained.
In this case, the Y coordinate representing the position of the common character in the search character string is corrected. In other words, the Y coordinate of “test” and “extraction”,
In addition, the Y coordinates of the “line” and the “go” are corrected so as to match.

Thus, the collation map generator 150 generates a Y coordinate value representing the position of the common character in the collation character string and a Y correction value which is a correction value of the position of the common character when the collation character string contains a synonym. A value, an X coordinate value representing the position of a common character in the collated character string, and a document identification number corresponding to the collated data are output as a collation map. Continuity evaluation unit 160
Inputs the collation map from the comprehensive map generation unit 150. The continuity evaluation unit 160 evaluates the similarity between the collated character string and the collated character string for each document identification number. Therefore, the continuity evaluation unit 160 first tracks the collation position in the collation map, assigns a continuity evaluation value to all collation positions, and then exists for characters in the same collation character string. The maximum value is selected as the continuity evaluation value of the character in the collation character string from the continuity evaluation values of the plurality of collation positions. Lastly, the continuity evaluation value obtained for each character in the collation character string is totaled and normalized for the entire collation character string, and the obtained value is used as the matching degree between the collation character string and the collated character string. The degree of match is
The continuity evaluation unit 160 sends the search result output unit 170 together with the document identification number.

Hereinafter, the continuity evaluation will be described in detail. FIG.
FIG. 4 is an explanatory diagram of path tracking for continuity evaluation performed in the character string collation system according to one embodiment of the present invention. In the route tracking processing, with respect to the collation positions in the collation map shown in FIG. 7, another collation position within an effective distance from one collation position is searched for and a link is established. By repeating this route tracking process, the matching position in the matching map is classified into several routes including branches. FIG. 9 shows a path from “high” to “fast”, and a “sentence” including a branch from “sho” to “no” and a branch from “sho” through “ken” to “search”. The path including the “book” is shown.

FIG. 10 is a diagram for explaining four typical types of continuity of the collation position. In general, where a continuous character string is collated, the link of the collation position is the lower right 45
Line up in the direction of degrees. FIG. 7A is a diagram showing a case where all the matching positions are completely matched in the lower right direction at 45 degrees. (B) of the figure shows a case where there is (one character) missing in data, (C) shows a case where there is (one character) replacement of data,
(D) is a diagram showing a case where data (two characters) is mixed. By tracking these links, the collation can be evaluated while maintaining continuity even when data is missing, replaced, or mixed.

The continuity evaluation unit 160 performs a process of calculating the degree of matching following the generation of the route. Here, in order to weight the link between the collation positions, a character type (type) is set for all the characters, and each character at the collation position is classified. In this example, two character types, “Kanji” and “Kana”, are used.
Assuming the case of classification into types, kanji: "high""fast"
"Sentence", "Book", "Check", "Kana": Kana is classified as "No". Next, according to the character types t1 and t2 of the characters before and after the link, the weight of the link between the character types is set as follows. Weight between character types (W _t ) = f (t ₁ , t ₂ ) = W11 (when t ₁ = Kanji, t ₂ = Kanji) = W12 (When t ₁ = Kanji, t ₂ = Kana) = W21 (T ₁ = Kana, t ₂ = Kanji) = W22 (T ₁ = Kana, t ₂ = Kana) Also, a weight is set for the link according to the length of the link (the distance between the collation positions before and after the link). For example, the weight according to the link length is expressed as follows. Weight by link length (W _l ) = g (x ₁ , y ₁ , x ₂ , y ₂ ) = 1 / {(x ₂ −x ₁ ) ² + (y ₂ −y ₁ ) ² } Finally, By combining the weight (W _t ) between the character types and the weight (W _l ) based on the length of the link, the following evaluation value for one link is obtained. 1 link evaluation value _{= v = W t · W l} = f (t 1, t 2) · g in _{(x 1, y 1, x} 2, y 2) continuity evaluation unit 160, then the lookup map For all the links on the route obtained by the route tracing, the evaluation values of the links are totaled, and the evaluation value of one entire route is obtained. The evaluation value V of this one route can be calculated according to the following formula, for example.

[0029]

(Equation 1) In the equation, k is the index of the link on the path of interest, n is the total number of links on the path of interest + 1, and v _k is the evaluation value of each link on the path of interest. The evaluation value V of one path thus obtained is set as an evaluation value _Vxy of the collation position at each collation position on the path of _interest . When the route includes a branch, for example, a route including a branch having the highest evaluation value among the evaluation values of the routes calculated for each branch can be selected as valid. In this way, by calculating the evaluation value V of the above one path for all the paths generated in the collation map, the evaluation value V _xy of the collation position is obtained for all the collation positions in the collation map. can get.

Next, an evaluation value for each character string in the collation character string is obtained. For example, as shown in FIG. 7, when there is at most one collation position corresponding to the character in the collation character string, the collation position is determined as the evaluation value of the character in the collation character string where the corresponding collation position exists. The evaluation value of the position is set, and the evaluation values of other characters in the collation character string are set to zero. If there are two or more collation positions corresponding to the character in the collation character string, the largest evaluation value among the evaluation values of the corresponding collation positions is set as the evaluation value of the character. Thus, the continuity evaluation value can be obtained for all the characters in the collation character string.

Finally, in order to determine the degree of matching of the entire collation character string with the collated character string, the continuity evaluation values of all the characters in the collation character string are totaled to obtain a total value. The _total value V _total of the continuity evaluation values is calculated, for example, according to the following equation.

[0032]

(Equation 2)For example, the degree of matching with the matched string as the entire matched string is
For example, the total value V of the continuity evaluation value_totalAn exact match of
Total value V of continuity evaluation value in case_equalBy the value divided by
Is expressed. Matching degree = V_total/ V_equal  By expressing the degree of coincidence in this way, the continuity evaluation value
Takes the maximum value of 1.0 in the case of a perfect match. this
The matching degree obtained in this way is referenced together with the document identification number.
The result is sent to the next search result output unit 170 as a combined result.

FIG. 11 is a configuration diagram of a search result output unit of the character string collating system according to one embodiment of the present invention. As shown in the figure, the search result output section 170 includes a collation result conversion section 171, a search result display section 172, a search result selection section 173, and a document display section 174. The matching result conversion unit 171 inputs the matching degree and the document identification number from the continuity evaluation unit 160 as the matching result, and based on the document identification number, searches the document heading, summary information, and the like corresponding to the matching result for the searched document. The information is read from the file 140, and the information on the document as the collation result is rearranged in the order of the matching degree, and is output as the retrieval result.

The search result display unit 172 inputs the search result from the collation result conversion unit 171, displays the search result on a display device such as a display, and passes the search result to the search result selection unit 173 in the next stage. The search result selection unit 173
A search result is input from the search result display unit 172, and an instruction from the operator corresponding to the search result display is input. When a document to be read is specified by the operator, the specified document is read from the search target document file 140. Loading,
Output as selected document.

The document display unit 174 includes the search result selection unit 17
The selected document output from 3 is input, and the read selected document is displayed on a display device such as a display. The character string matching system according to one embodiment of the present invention is shown in FIGS.
In accordance with the configuration and operation described with reference to 1, the search text input by the operator is compared with the document stored in the search target document file, and a document including the search text similar to the search text is presented to the operator. Can be.

Next, in the character string collating system according to the embodiment of the present invention, a process when the search sentence expanding unit 111 outputs an extended search sentence will be described. In this example, it is assumed that a synonym “extraction” of the character string “search” exists in the search sentence “speed-up of document search”. As described above, when the collation data includes a synonym, it is interpreted that there are a plurality of collation data “speed-up document search” and “speed-up document extraction”. In addition, by using the synonym data regular expression: speeding up document {search | extraction}, an extended search sentence is created in which synonyms are enumerated in the collation data. When the collation data includes a synonym, the collation data is processed as if one of the synonyms at the same position is selected. FIG. 12 is an explanatory diagram of the path tracking in the matching map including the synonyms shown in FIG. The effective distance at the time of route tracking is the distance between the matching positions on the route actually arranged in the matching map generated, and the theoretical distance generated when one synonym is selected and processed. It is calculated in consideration of a distance correction value indicating a difference from a distance between collation positions on the collation map.

Lastly, a description will be given of a process in the case where the numerical expression in the collation character string or the collated character string is replaced with a numerical value in the numerical expression replacement unit 113 or 133 of the character string collation system according to one embodiment of the present invention. I do. FIG.
It is a flowchart of the processing procedure of similar quantitative character collation.
First, a partial character string quantitatively expressed by a numerical value is extracted from the collation character string or the collated character string (step 10).
0). Second, the extracted partial character string is converted into a value (step 101). Third, the degree of matching of numerical values is calculated based on the converted values (step 102).

Here, the extraction of a partial character string quantitatively expressed by a numerical value is performed by detecting and extracting a portion where a numerically expressed character continuously appears in the character string. For example, the following characters are detected as numerical expression characters. 1 2 3 4 5 6 7 8 9 0 1 2 4 5 6 7 8 7 9 9 0 10 million ... The degree of matching is calculated from the value (Vs) obtained from the collation character string and the collated character string. It can be calculated according to the following equation based on the obtained value (Vd).

[0039]

(Equation 3) FIG. 14 is an explanatory diagram of the similar quantitative character collation processing. The figure shows a collation character string and a collation target character string, a partial character string by a quantitative expression, a numerical value converted from the partial character string, and a degree of matching between the converted numerical values. Therefore, according to an embodiment of the present invention, in a character string collating system using a computer, a partial character string is missing in a collated character string or a collated character string, replacement with another character string, another character string. When character strings that partially match are scattered due to mixing of characters, character string collation can be performed.

Further, the configuration of the character string collating system according to one embodiment of the present invention is not limited to the example described in the above embodiment, and each component of the character string collating system is defined by software (program). It is also possible to record the data in a disk device or the like, install it in a computer of a character string collating system as needed, and perform character string collation. Furthermore, the constructed program can be stored in a portable recording medium such as a floppy (registered trademark) disk or CD-ROM, and can be used for general purposes in a case where such a character string collation system is used.

The present invention is not limited to the above embodiment, but can be variously modified and applied within the scope of the claims.

[0042]

As described above, according to the present invention, when performing pattern matching, the pattern matching position can be tracked, and the continuity of the pattern can be evaluated even if the matching positions are separated. Therefore, a part of the matching pattern or the to-be-checked pattern is missing, replaced with another pattern, or mixed with another pattern. Can be collated even if there are patterns that are distributed. Therefore, according to the present invention, even if the operator is not familiar with the content of the pattern to be verified, verification without omission is realized, and an advantage that the burden on the operator is reduced is obtained.

[Brief description of the drawings]

FIG. 1 is a principle configuration diagram of the present invention.

FIG. 2 is an operation flowchart of the similarity information matching method of the present invention.

FIG. 3 is a schematic configuration diagram of a character string collation system according to an embodiment of the present invention.

FIG. 4 is an operation flowchart of the character string collation system according to one embodiment of the present invention.

FIG. 5 is a configuration diagram of a collation data generation unit of the character string collation system according to one embodiment of the present invention.

FIG. 6 is a configuration diagram of a collated data generation unit of the character string collation system according to one embodiment of the present invention.

FIG. 7 is an explanatory diagram visually representing a matching map according to an embodiment of the present invention.

FIG. 8 is an explanatory diagram of a matching map when a synonym is included.

FIG. 9 is an explanatory diagram of path tracking for continuity evaluation performed in the character string collating system according to one embodiment of the present invention.

FIG. 10 is a diagram for explaining the continuity of collation positions.

FIG. 11 is a configuration diagram of a search result output unit of the character string collation system according to one embodiment of the present invention.

FIG. 12 is an explanatory diagram of route tracking in a collation map including the synonyms shown in FIG. 8;

FIG. 13 is a flowchart of a processing procedure for similar quantitative character matching.

FIG. 14 is an explanatory diagram of a similar quantitative character collation process.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Similar information collation apparatus 10 Pattern generation means 20 Collation map creation means 30 Collation map 40 Continuity evaluation means 50 Pattern collation means

Claims (11)

[Claims] 1. A similarity information matching device for determining the similarity of information, wherein the first information and the second information to be matched are used as a pattern represented by a position and a characteristic of an element of the information. Pattern generating means for generating a first pattern corresponding to the first information and a second pattern corresponding to the second information, and a second pattern having the same feature in the first pattern and the second pattern. A collation map generating means for producing a collation map composed of a collation position having coordinates of pairs of positions of a first element belonging to one pattern and a second element belonging to the second pattern; A continuity evaluation means for evaluating the continuity of the route for each of the paths in which the matching positions in the vicinity are sequentially connected, and the first pattern based on the continuity evaluated for each of the routes. The matching of the emission and the second pattern and a determining pattern matching means, similarity information collating apparatus. 2. The collation map creating means creates the collation positions individually for a plurality of combinations of the first element and the second element having the same characteristics. Described similarity collation device. 3. The pattern matching means sets, for each of the matching positions, the highest continuity among continuities evaluated for the path passing through the matching position as an evaluation value of the matching position. Based on the evaluation value set for each of the collation positions,
3. The similarity information matching device according to claim 1, further comprising: means for calculating a degree of coincidence between the second pattern and the second pattern. 4. The pattern generating means according to claim 1, wherein at least a part of the information represented as the pattern is replaced with a synonymous element having a characteristic capable of replacing a characteristic of the at least part of the original element. Generating means; and means for generating the pattern so that the synonymous elements are listed at the same time as the original element. The matching map generating means, the continuity evaluating means, and the pattern matching means 4. The similarity matching apparatus according to claim 1, wherein the synonymous elements listed at the same time are adapted to be processed in parallel with the original elements. . 5. The method according to claim 1, wherein said collation map generating means has means for judging that said element has the same characteristic when said element has a characteristic expressing a numerical value, when said element has a characteristic expressing a numerical value. Item 5. The similarity information matching device according to any one of Items 1 to 4. 6. A similarity information matching method for determining similarity of information by comparing a first pattern and a second pattern represented by positions and characteristics of information elements, wherein the first pattern and the Inputting a second pattern; and a first element belonging to the first pattern and a second element belonging to the second pattern having the same characteristics among the first pattern and the second pattern. Detecting an element; generating a verification map having coordinates of pairs of the detected positions of the first element and the second element; and generating a verification map having coordinates in the vicinity of the verification map. A path generation step of generating a path by sequentially connecting collation positions; a continuity evaluation step of evaluating the continuity of the path for each of the generated paths; and a continuity evaluated for each of the paths. A pattern matching step of determining a degree of matching between the first pattern and the second pattern based on the pattern matching step. 7. A computer-readable recording medium having recorded thereon a similarity information collation program for judging the similarity of information, wherein the similarity information collation program comprises: a first information and a second information to be collated; A pattern generation code for generating a first pattern corresponding to the first information and a second pattern corresponding to the second information as the pattern represented by the position and the characteristic of the element of the information; And a collation position having coordinates of a pair of positions of a first element belonging to the first pattern and a second element belonging to the second pattern having the same feature in the second pattern and the second pattern. A collation map generating code for creating a collation map composed of: Recording medium, comprising: a continuity evaluation code for evaluating the continuity of a pattern; and a pattern matching code for determining a degree of matching between the first pattern and the second pattern based on the continuity evaluated for each path. . 8. The similarity information collation method according to claim 1, wherein the collation map creation code causes the collation position to be created individually for a plurality of combinations of the first element and the second element having the same characteristics. The recording medium according to claim 7, wherein the program is recorded. 9. A code for setting the highest continuity among the continuities evaluated for the route passing through the matching position for each of the matching positions as an evaluation value of the matching position for each of the matching positions. Based on the evaluation value set for each of the collation positions,
9. The recording medium according to claim 7, wherein a similarity information matching program having a pattern for calculating the degree of matching between the second pattern and the second pattern is recorded. 10. The pattern generation code according to claim 1, wherein at least a part of the information represented as the pattern has a synonymous element having a characteristic capable of replacing a characteristic of the at least some original element. And a code for generating the pattern so that the synonymous elements are enumerated at the same time as the original element. The synonymous elements enumerated at the same time are in parallel with the original element. The recording medium according to any one of claims 7 to 9, wherein a similarity information collation program is adapted to be processed. 11. The similarity characteristic that the collation map generation code has a code that, when the element has a characteristic expressing a numerical value, determines that the element has the same characteristic when the value represented by the numerical value matches. The recording medium according to any one of claims 7 to 10, wherein the information collation program is recorded.