JP2005004626A - Different notation normalization processor, different notation normalization processing program, storage medium storing the same, document retrieval device, document retrieval program and storage medium storing the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a different notation normalization processor by which fluctuation in various notations among various languages having different notations in Chinese character is absorbed. <P>SOLUTION: The different notation normalization processor comprises: a document input means 101 receiving an input text; a Chinese character equivalence segmentation means 102a extracting the notation in Chinese character from the input text received by the document input means 101; at least two or more of different notation normalization rules 102c describing corresponding characters among the languages having the different notations in Chinese character; a different notation normalization rule selection means 102d selecting the different notation normalization rule 102c suitable for the input text received by the document input means 101; and a normalization processing means 102b normalizing the notation in Chinese character extracted by the Chinese character equivalence segmentation means 102a in accordance with the different notation normalization rule 102c selected by the different notation normalization rule selection means 102d. In this way, the fluctuation in the various notations among the various languages having the different notations in Chinese character is absorbed. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a different notation normalization processing apparatus, a different notation normalization processing program, a storage medium storing the same, a document search apparatus, a document search program, and a storage medium storing the same.
[0002]
[Prior art]
Chinese characters commonly used in China are called simplified characters, Chinese characters commonly used in Taiwan are called traditional characters, and different Chinese characters may be used to represent the same meaning. For example, what is represented as “printing machine” in traditional Chinese characters is represented as “printing machine” in simplified Chinese characters.
[0003]
Therefore, an apparatus for converting a document written in simplified characters and a document written in traditional characters has been researched and developed (for example, see Patent Document 1).
[0004]
[Patent Document 1]
JP-A-8-263478
[0005]
[Problems to be solved by the invention]
By the way, examples of languages that use Chinese characters include Japanese, Korean, Vietnamese, etc. in addition to the simplified and traditional Chinese characters described above. In these languages, the pronunciation of kanji is different but sometimes used in the same meaning, and can be understood with simple sentences. Even if the languages are the same, there are old characters, new characters, and variant characters in each language, and different kanji characters may be used to represent the same meaning.
[0006]
As described above, when a document search process is performed on a document in which different kanji characters are used to express the same meaning, there is a problem that a search omission occurs due to different kanji notations.
[0007]
For this reason, in recent years, it has been desired to absorb various notation fluctuations between various languages as well as between simplified and traditional characters.
[0008]
An object of the present invention is to provide a different notation normalization processing apparatus, a different notation normalization processing program, and a storage medium for storing the same, which can absorb various notation fluctuations between various languages having different kanji notations. It is.
[0009]
An object of the present invention is to prevent occurrence of a search omission in a document search process for a document written in a language in a Chinese-speaking area (Chinese (simplified or traditional), Japanese, Korean, Vietnamese, etc.) To provide a document search apparatus, a document search program, and a storage medium for storing the document search program capable of searching the most appropriate document.
[0010]
[Means for Solving the Problems]
The different notation normalization processing apparatus according to claim 1 is a document input means for receiving input text, a kanji equivalent cutout means for extracting kanji notation from the input text received by the document input means, and kanji notation. Selection of normalization rule storage means for storing at least two kinds of different notation normalization rules describing character correspondence between different languages, and the different notation normalization rules suitable for the input text received by the document input means And a normalization processing means for normalizing the kanji notation extracted by the kanji equivalent cutout means according to the different notation normalization rule selected by the different notation normalization rule selecting means. .
[0011]
Therefore, the kanji notation is extracted from the input text, and the different notation normalization rule suitable for the input text is selected from at least two different notation normalization rules describing the character correspondence between languages having different kanji notations. The kanji notation extracted based on the different notation normalization rules is normalized. As a result, it is possible to absorb various notation fluctuations between various languages having different kanji notations.
[0012]
The invention according to claim 2 is the different notation normalization processing device according to claim 1, wherein the different notation normalization rule selection unit analyzes language information of the input text, and based on the analyzed language information, Select a different notation normalization rule.
[0013]
Therefore, it is possible to perform an appropriate different notation normalization process for the input text in kanji without being conscious of which language is used for input.
[0014]
According to a third aspect of the present invention, in the different notation normalization processing device according to the second aspect, the different notation normalization rule selecting unit selects a plurality of the different notation normalization rules based on the analyzed language information. In such a case, it is allowed to select a desired different notation normalization rule from these different notation normalization rules.
[0015]
Therefore, it is possible to perform the most appropriate different notation normalization process on the input text of kanji.
[0016]
The invention according to claim 4 is the different notation normalization processing device according to any one of claims 1 to 3, wherein the different notation normalization rule is a character between a kanji notation of a predetermined language and its variant character. The correspondence is described.
[0017]
Therefore, it becomes possible to absorb the fluctuation of the notation between the kanji notation of a predetermined language and its variant characters.
[0018]
According to a fifth aspect of the present invention, in the different notation normalization processing device according to any one of the first to third aspects, the different notation normalization rule is a kanji notation of another language corresponding to a kanji notation of a predetermined language. Is described.
[0019]
Therefore, it becomes possible to absorb the fluctuation of the notation between the kanji notation of a predetermined language and the kanji notation of another language.
[0020]
The invention according to claim 6 is the variant notation normalization processing device according to any one of claims 1 to 3, wherein the variant notation in which the character correspondence between the kanji notation of the predetermined language and the variant is described. Both the normalization rule and the different notation normalization rule in which the kanji notation of another language corresponding to the kanji notation of a predetermined language is described.
[0021]
Therefore, it becomes possible to absorb the fluctuation of various notations between various languages having different kanji notations.
[0022]
A different notation normalization processing program according to claim 7 is installed in a computer and has a document input function for receiving input text, and a kanji equivalent cutout function for extracting kanji notation from the input text received by the document input function. A normalization rule storage function for storing at least two kinds of different notation normalization rules describing character correspondence between languages having different kanji notations, and the different notation suitable for the input text received by the document input function A normalization rule selection function for selecting a normalization rule, and a Chinese character notation extracted by the kanji equivalent cutout function is normalized according to the normalization normalization rule selected by the normalization normalization rule selection function. Let the computer execute the normalization processing function.
[0023]
Therefore, the kanji notation is extracted from the input text, and the different notation normalization rule suitable for the input text is selected from at least two different notation normalization rules describing the character correspondence between languages having different kanji notations. The kanji notation extracted based on the different notation normalization rules is normalized. As a result, it is possible to absorb various notation fluctuations between various languages having different kanji notations.
[0024]
The invention according to claim 8 is the different notation normalization processing program according to claim 7, wherein the different notation normalization rule selection function analyzes language information of the input text, and based on the analyzed language information, Select a different notation normalization rule.
[0025]
Therefore, it is possible to perform an appropriate different notation normalization process for the input text in kanji without being conscious of which language is used for input.
[0026]
The invention according to claim 9 is the different notation normalization processing program according to claim 8, wherein the different notation normalization rule selection function selects a plurality of the different notation normalization rules based on the analyzed linguistic information. In such a case, it is allowed to select a desired different notation normalization rule from these different notation normalization rules.
[0027]
Therefore, it is possible to perform the most appropriate different notation normalization process on the input text of kanji.
[0028]
The invention according to claim 10 is the variant notation normalization processing program according to any one of claims 7 to 9, wherein the variant notation normalization rule is a character between a kanji notation of a predetermined language and its variant character. The correspondence is described.
[0029]
Therefore, it becomes possible to absorb the fluctuation of the notation between the kanji notation of a predetermined language and its variant characters.
[0030]
The invention according to claim 11 is the different notation normalization processing program according to any one of claims 7 to 9, wherein the different notation normalization rule is a kanji notation of another language corresponding to a kanji notation of a predetermined language. Is described.
[0031]
Therefore, it becomes possible to absorb the fluctuation of the notation between the kanji notation of a predetermined language and the kanji notation of another language.
[0032]
The invention according to claim 12 is the variant notation normalization processing program according to any one of claims 7 to 9, wherein the variant notation in which character correspondence between the kanji notation of a predetermined language and its variant is described. Both the normalization rule and the different notation normalization rule in which the kanji notation of another language corresponding to the kanji notation of a predetermined language is described.
[0033]
Therefore, it becomes possible to absorb the fluctuation of various notations between various languages having different kanji notations.
[0034]
A storage medium according to a thirteenth aspect of the invention stores the different notation normalization processing program according to any one of the seventh to twelfth aspects of the invention.
[0035]
Therefore, by causing the computer to read the different notation normalization processing program stored in the storage medium, it is possible to obtain the same operation as that of the invention according to any one of claims 7 to 12.
[0036]
A document retrieval apparatus according to a fourteenth aspect of the present invention is a different notation normalization processing apparatus according to any one of claims 1 to 6 that performs different notation normalization processing, a document input unit that receives input text, and the document input If the input text received by the means is a document, the document storage means for storing in the document database, and the character storage is extracted from the document stored in the document database by the document storage means, and the different notation for each extracted character string After the normalization processing is performed by the different notation normalization processing device, index registration means for extracting index information and storing it in the index information storage unit, and when the input text received by the document input means is a search word, The search term is sent to the different notation normalization processing device, a character string is extracted from the search term, and the different notation normalization processing is performed. Search condition creating means for creating a search result, search processing means for retrieving index information in the index information storage unit based on the search condition created by the search condition creating means, and a result of the search by the search processing means. And a result output means for outputting.
[0037]
Therefore, index information and search conditions that have been subjected to different notation normalization processing are generated, so they are written in the language of the Chinese character use area (Chinese (simplified and traditional), Japanese, Korean, Vietnamese, etc.). Since it is possible to prevent the occurrence of a search omission in the document search process for the selected document, it is possible to search for the most appropriate document.
[0038]
A document search program according to a fifteenth aspect of the invention is installed in a computer and has a different notation normalization processing function for realizing a different notation normalization process in the different notation normalization processing device according to any one of claims 1 to 6. A document input function that accepts an input text, a document storage function that stores the input text received by the document input function in a document database, and a document that is stored in the document database by the document storage function. An index registration function for extracting the index information and storing it in the index information storage unit after extracting the character string and performing the different notation normalization processing for each extracted character string with the different notation normalization processing function, and the document input When the input text received by the function is a search word, the search word is transferred to the different notation normalization processing function. A search condition creating function for creating a search condition after extracting a character string from the search term and performing normal notation processing, and the index based on the search condition created by the search condition creating function A computer is caused to execute a search processing function for searching index information in the information storage unit and a result output function for outputting a search result by the search processing function.
[0039]
Therefore, index information and search conditions that have been subjected to different notation normalization processing are generated, so they are written in the language of the Chinese character use area (Chinese (simplified and traditional), Japanese, Korean, Vietnamese, etc.). Since it is possible to prevent the occurrence of a search omission in the document search process for the selected document, it is possible to search for the most appropriate document.
[0040]
A storage medium according to a sixteenth aspect stores the program according to the fifteenth aspect.
[0041]
Therefore, by causing the computer to read the document search program stored in the storage medium, it is possible to obtain the same operation as that of the invention of the fifteenth aspect.
[0042]
DETAILED DESCRIPTION OF THE INVENTION
A first embodiment of the present invention will be described with reference to FIGS.
[0043]
FIG. 1 is a block diagram schematically showing a hardware configuration of a different notation normalization processing apparatus 1 to which the present invention is applied. As shown in FIG. 1, the different notation normalization processing apparatus 1 is, for example, a personal computer or a workstation, and includes a CPU (Central Processing Unit) 2 that is a main part of the computer and controls each part centrally. . The CPU 2 is connected by a bus 5 to a ROM (Read Only Memory) 3 which is a read only memory storing BIOS and a RAM (Random Access Memory) 4 which stores various data in a rewritable manner.
[0044]
Further, the bus 5 has an HDD (Hard Disk Drive) 6 for storing various programs and the like, and a CD-ROM drive 8 for reading a CD (Compact Disc) -ROM 7 as a mechanism for reading computer software as a distributed program. A communication control device 10 that controls communication between the different notation normalization processing device 1 and the network 9, an input device 11 such as a keyboard and a mouse, and a display device such as a CRT (Cathode Ray Tube) and an LCD (Liquid Crystal Display). 12 are connected via an I / O (not shown).
[0045]
Since the RAM 4 has a property of storing various data in a rewritable manner, the RAM 4 functions as a work area for the CPU 2 and plays a role of, for example, a document buffer described later.
[0046]
A CD-ROM 7 shown in FIG. 1 implements the storage medium of the present invention, and stores an OS (Operating System) and various programs. The CPU 2 reads the program stored in the CD-ROM 7 with the CD-ROM drive 8 and installs it in the HDD 6.
[0047]
As the storage medium, not only the CD-ROM 7 but also various types of media such as semiconductor memory such as various optical disks such as DVD, various magnetic disks such as various magneto-optical disks and flexible disks, and the like can be used. Alternatively, the program may be downloaded from the network 9 such as the Internet via the communication control device 10 and installed in the HDD 6. In this case, the storage device storing the program in the server on the transmission side is also a storage medium of the present invention. Note that the program may operate on a predetermined OS (Operating System), and in that case, the OS may take over the execution of some of the various processes described below, It may be included as a part of a group of program files constituting the application software or OS.
[0048]
The CPU 2 that controls the operation of the entire system executes various processes based on a program loaded on the HDD 6 used as the main storage of the system.
[0049]
Next, the contents of various processes executed by the CPU 2 of the different notation normalization processing apparatus 1 based on a program will be described. FIG. 2 is a functional block diagram showing the different notation normalization processing apparatus 1. As shown in FIG. 2, in the different notation normalization processing apparatus 1, the document input means 101 that accepts text (input text) such as document data or a search term input from the input device 11, and the input text is normalized. Each function of the output notation normalizing means 102 for performing the normalization processing, and the output means 103 for outputting the processing results such as information related to the normalized notation output from the different notation normalizing means 102 to the display device 12, the CPU 2 is a computer It is realized by operating according to software. When used for a document search or the like, it is also possible to directly output the processing result in the different notation normalization unit 102 to the indexing process or the search process without using the output unit 103. Further, the input of text such as document data and search terms is not limited to the input from the input device 11, but may be input by OCR (Optical Character Reader) or the like.
[0050]
Here, the different notation normalization means 102 will be described in detail. As shown in FIG. 2, the different notation normalizing means 102 includes a kanji equivalent cutout means 102a, a normalization processing means 102b, and a normalization rule storage means 102e for storing at least two kinds of different notation normalization rules 102c. And a different notation normalization rule selection means 102d.
[0051]
The kanji equivalent cutout means 102a extracts characters in Chinese characters one by one from the text (input text) input via the document input means 101 and outputs them to the normalization processing means 102b.
[0052]
The different notation normalization rule 102c stored in the normalization rule storage unit 102e generally describes character correspondence between predetermined languages. FIG. 3 is an explanatory diagram showing an example of the different notation normalization rule 102c. The different notation normalization rule 102c shown in FIG. 3 is a different notation normalization rule for conversion between traditional Chinese characters and simplified Chinese characters. As shown in FIG. 3, the traditional Chinese character notation and character code (Unicode) and the simplified Chinese character notation and character code (Unicode) are associated with the different notation normalization rule 102 c. As shown in FIG. 4, the different notation normalization rule 102c may describe different notation normalization rules between a plurality of languages, and the character code may be a character code other than Unicode, such as JIS0208 or UTF-8. Multiple types of character codes may be described. Thereby, it becomes possible to absorb the fluctuation of the notation between the kanji notation of a predetermined language and the kanji notation of another language. Further, the different notation normalization rule 102c may be a different notation normalization rule for conversion between language variants and traditional characters, as shown in FIG. Thereby, it becomes possible to absorb the fluctuation of the notation between the kanji notation of a predetermined language and its variant characters. Such a different notation normalization rule 102c is selected by the different notation normalization rule selection means 102d.
[0053]
The different notation normalization rule selection unit 102d analyzes the language information of the document (input text) input from the input device 11 and received by the document input unit 101, and the different notation normalization rule 102c based on the analyzed language information. Is extracted from the normalization rule storage means 102e. As the language information of the document, information in which the language information of the document is explicitly input separately from the document from the input device 11 or language information described in advance in the input document is used. For example, in an XML document, language information is described using a lang attribute. As a method for discriminating such language information, an existing language identification method (for example, a method disclosed in JP-A-10-320399) can be used. Therefore, it is possible to perform an appropriate different notation normalization process for the input text in kanji without being conscious of which language is used for input. When a plurality of different notation normalization rules 102c are extracted, a list of different notation normalization rules 102c extracted is displayed on the display device 12, and an appropriate different notation normalization rule is selected from the list. 102c may be selected by the input device 11. As a result, the most appropriate different notation normalization process can be performed on the input text in kanji.
[0054]
The normalization processing unit 102b generates a normalized notation based on the different notation normalization rule 102c selected by the different notation normalization rule selection unit 102d for the single kanji character output from the kanji equivalent cutout unit 102a. To do.
[0055]
Next, the overall flow of the different notation normalization process in the different notation normalization processing apparatus 1 will be described with reference to FIG. In FIG. 6, the Chinese traditional Chinese-Simplified Chinese conversion (see FIG. 3) is selected by the different notation normalization rule selection means 102 d as the different notation normalization rule 102 c, and “printing machine” is selected as the input text in the traditional Chinese character. Is shown as an example.
[0056]
As shown in FIG. 6, first, text input via the document input means 101 is received by the document input means 101 (Y in step S1), and is included in the text input by the kanji equivalent cutout means 102a. One kanji character is detected and cut out (step S2). Here, first, the Chinese character “mark” is cut out.
[0057]
The extracted Chinese character “mark” is input to the normalization processing means 102b, and normalization processing is performed based on the different notation normalization rule 102c (step S3). The normalization process confirms whether or not the extracted Chinese characters conform to the different notation normalization rules 102c (Traditional Chinese characters-Simplified Chinese conversion), and the different notation normalization rules 102c (Traditional Chinese characters-Chinese characters). When conforming to (simplified Chinese character conversion), normalization is performed according to the different notation normalization rule 102c (traditional Chinese character-Simplified Chinese character conversion). Here, since there is no rule corresponding to the different notation normalization rule 102c (traditional Chinese character-Simplified Chinese character conversion) shown in FIG. 3 for the cut out Chinese character “mark”, the process directly proceeds to step S4.
[0058]
Since all the kanji characters included in the input text have not been cut out (N in step S4), the process returns to step S2, and the next character “print” is cut out by the kanji equivalent cutting means 102a.
[0059]
Next, the kanji “print” cut out does not have a normalization rule because there is no rule corresponding to the different notation normalization rule 102c (traditional Chinese character-simplified Chinese conversion) shown in FIG. S3) Proceed directly to step S4. Then, since all the kanji characters included in the input text have not been cut out (N in step S4), the process returns to step S2, and the next character “machine” is cut out by the kanji equivalent cutting means 102a.
[0060]
The next extracted Chinese character “machine” has a rule corresponding to the different notation normalization rule 102c (Traditional Chinese character-Simplified Chinese character conversion) shown in FIG. The character “machine” is normalized to the Chinese simplified character “desk” (step S3). Proceed to step S4. Then, since all the kanji characters included in the input text have not been cut out (N in step S4), the process returns to step S2, and the next character “machine” is cut out by the kanji equivalent cutting means 102a.
[0061]
Next, the kanji character “machine” cut out does not have normalization because there is no rule corresponding to the different notation normalization rule 102c (traditional Chinese character-simplified Chinese character conversion) shown in FIG. S3) Proceed directly to step S4. Since all the kanji characters included in the input text have been cut out (Y in step S4), the process proceeds to step S5.
[0062]
Therefore, the Chinese character string “printing machine”, which is the traditional Chinese character in the character string of the original text, is changed to the simplified Chinese character “printing machine” in accordance with the different notation normalization rule 102c (Traditional Chinese-Simplified Chinese character conversion). Converted. The character string “printing machine” subjected to normalization in this way is output to the display device 12 as a processing result of the normalization processing means 102b (step S5).
[0063]
As described above, according to the present embodiment, kanji notation is extracted from the input text, and the different notation normalization rule 102c describing the character correspondence between languages having different kanji notations is used. The notation normalization rule 102c is selected, and the kanji notation extracted based on the selected different notation normalization rule 102c is normalized. As a result, it is possible to absorb various notation fluctuations between various languages having different kanji notations.
[0064]
In this embodiment, the character string to be normalized is described as an example of conversion between traditional Chinese characters and simplified Chinese characters. However, the present invention is not limited to this. It is possible to apply to the case where the same notation normalization processing is performed between countries where Kanji is used, such as in the case of Vietnamese Kanji and Simplified Chinese.
[0065]
Further, the different notation normalization rule selection means 102d converts both the traditional Chinese character-Simplified Chinese character conversion (see FIG. 3) and the variant language character-Chinese traditional character character conversion (see FIG. 5) as the different character normalization rule 102c. For example, if the input Chinese character is the language variant “country”, it is normalized to the Chinese traditional character “country” according to the language variant-Chinese traditional character conversion (see FIG. 5), and further, the Chinese traditional character— The traditional Chinese character “country” may be normalized to the simplified Chinese character “country” according to the simplified Chinese character conversion (see FIG. 3).
[0066]
Next, a second embodiment of the present invention will be described with reference to FIGS. The same parts as those described in the first embodiment of the present invention are denoted by the same reference numerals, and description thereof is also omitted. The present embodiment relates to a document search apparatus provided with the different notation normalization processing apparatus 1 of the first embodiment.
[0067]
FIG. 7 is a functional block diagram showing a schematic configuration of the document search device 50 according to the present embodiment, FIG. 8 shows a processing flow in the document search device 50, and (a) extracts index information from a character string included in the document. FIG. 5B is a flowchart showing an outline of a flow of document search processing for searching for a document using a search word.
[0068]
The document retrieval apparatus 50 of this embodiment is a personal computer or workstation similar to the different notation normalization processing apparatus 1 described in the first embodiment, and has a hardware configuration that is different from the different notation normalization processing apparatus. 1 is the same. That is, the document retrieval apparatus 50 is different from the different notation normalization processing apparatus 1 in that a document retrieval program is loaded on the HDD 6.
[0069]
In the following, processing executed by the CPU 2 based on the document search program will be described along the functional block diagram shown in FIG. 7 and the processing flow shown in FIG.
[0070]
First, the flow of index information registration processing will be described. The document storage unit 53 reads a document input from the input device 51 such as a keyboard and a mouse, the input device 51 such as an OCR, and the like through the document input unit 52, and sequentially registers the read document in the document database 54 ( Step S11).
[0071]
Next, if there is a document for which index information has not yet been extracted from among the registered documents, the index registration means 55 reads it from the document database 54 (step S12), and the document is stored in the different notation normalization processing device 1. In succession, a character string is extracted from the document, and a different notation normalization process is performed for each extracted character string (step S13). Since the different notation normalization processing has been described in the first embodiment, the description thereof is omitted here.
[0072]
Since the normalized notation information normalized in step S13 is returned to the index registration means 55, it is stored and saved in the index information storage unit 56 in a format (index information) associated with the document ( Step S14).
[0073]
Next, the flow of document search processing will be described. The search term input from the input device 51 is read by the search condition creation unit 57 via the document input unit 52, and the search condition creation unit 57 sends the search term to the different notation normalization processing device 1 to transmit the search term. A character string is extracted from the search term, and a different notation normalization process is performed (step S21). Since the different notation normalization processing has been described in the first embodiment, the description thereof is omitted here.
[0074]
Subsequently, the search condition creating unit 57 creates a search condition based on the normalized notation information subjected to the different notation normalization process, and the search processing unit 58 uses the index information stored and saved in the index information storage unit 56. The document search process is performed (step S22).
[0075]
If necessary, the document search result is read from the document database 54 and displayed on the display device 12 or the output device 59 such as the HDD 6 that outputs and saves the processing result (step). S23: Result output means).
[0076]
As described above, according to the present embodiment, index information and search conditions subjected to the different notation normalization processing are generated, so that the language of the kanji use range (Chinese (simplified and traditional), Japanese, Korean) Occurrence of a search omission in a document search process for a document described in Japanese, Vietnamese, etc.) can be prevented, so that the most appropriate document can be searched.
[0077]
【The invention's effect】
According to the different notation normalization processing apparatus of claim 1, the document input means for receiving the input text, the kanji equivalent extracting means for extracting the kanji notation from the input text received by the document input means, and the kanji Normalization rule storage means for storing at least two different notation normalization rules describing character correspondence between languages with different notations, and the different notation normalization rules suitable for the input text received by the document input means Normalization processing for normalizing the kanji notation extracted by the kanji equivalent cutout means according to the different notation normalization rule selected by the different notation normalization rule selecting means Means for extracting kanji notation from the input text, and describing at least two different notation normalization rules describing character correspondence between languages having different kanji notations By selecting different notation normalization rules suitable for the input text, and normalizing the extracted kanji notations based on the selected different notation normalization rules, a variety of languages between different languages with different kanji notations can be used. Can absorb the fluctuation of the notation.
[0078]
According to a second aspect of the present invention, in the different notation normalization processing device according to the first aspect, according to the different notation normalization rule selecting means, the language information of the input text is analyzed, and based on the analyzed language information. By selecting the different notation normalization rule, appropriate different notation normalization processing can be performed on the input text of kanji without being conscious of which language is used for input.
[0079]
According to a third aspect of the present invention, in the different notation normalization processing device according to the second aspect, the different notation normalization rule selecting means may include a plurality of the different notation normalization rules based on the analyzed language information. If selected, the most appropriate different notation normalization process is performed on the input text of the Chinese character by allowing selection of the desired different notation normalization rule from these different notation normalization rules. be able to.
[0080]
According to a fourth aspect of the present invention, in the different notation normalization processing device according to any one of the first to third aspects, the different notation normalization rule is a kanji notation of a predetermined language and its variant characters. By describing the correspondence between the characters, it is possible to absorb the fluctuation of the notation between the kanji notation of the predetermined language and its variant characters.
[0081]
According to a fifth aspect of the present invention, in the different notation normalization processing device according to any one of the first to third aspects, the different notation normalization rule may be set in another language corresponding to the kanji notation of a predetermined language. By describing the kanji notation, it is possible to absorb the fluctuation of the notation between the kanji notation of a predetermined language and the kanji notation of another language.
[0082]
According to the invention described in claim 6, in the different notation normalization processing device according to any one of claims 1 to 3, the character correspondence between the kanji notation of a predetermined language and the variant character is described. Various languages with different kanji notations by having both the different notation normalization rules and the different notation normalization rules in which kanji notations of other languages corresponding to the kanji notation of a given language are described It can absorb the fluctuations of various notations.
[0083]
According to the different notation normalization processing program of the invention described in claim 7, a document input function that is installed in a computer and receives input text, and a kanji equivalent cutout that extracts kanji notation from the input text received by the document input function. An output function, a normalization rule storage function for storing at least two different notation normalization rules describing character correspondence between languages having different kanji notations, and the input text received by the document input function. The different notation normalization rule selection function for selecting the different notation normalization rule and the kanji notation extracted by the kanji equivalent cutout function according to the different notation normalization rule selected by the different notation normalization rule selection function are normalized. The normalization processing function to convert to Kanji is extracted from the input text, and the Kanji notation is different. Select a different notation normalization rule suitable for the input text from at least two different notation normalization rules that describe the correspondence between languages, and extract the kanji notation extracted based on the selected different notation normalization rule By normalizing, it is possible to absorb various notation fluctuations between various languages having different kanji notations.
[0084]
According to an eighth aspect of the present invention, in the different notation normalization processing program according to the seventh aspect, the different notation normalization rule selection function analyzes language information of the input text, and based on the analyzed language information. By selecting the different notation normalization rule, appropriate different notation normalization processing can be performed on the input text of kanji without being conscious of which language is used for input.
[0085]
According to a ninth aspect of the present invention, in the different notation normalization processing program according to the eighth aspect, the different notation normalization rule selection function may include a plurality of the different notation normalization rules based on the analyzed language information. If selected, the most appropriate different notation normalization process is performed on the input text of the Chinese character by allowing selection of the desired different notation normalization rule from these different notation normalization rules. be able to.
[0086]
According to a tenth aspect of the present invention, in the different notation normalization processing program according to any one of the seventh to ninth aspects, the different notation normalization rule is a kanji notation of a predetermined language and its variant characters. By describing the correspondence between the characters, it is possible to absorb the fluctuation of the notation between the kanji notation of the predetermined language and its variant characters.
[0087]
According to an eleventh aspect of the present invention, in the different notation normalization processing program according to any one of the seventh to ninth aspects, the different notation normalization rule may be set in another language corresponding to the kanji notation of a predetermined language. By describing the kanji notation, it is possible to absorb the fluctuation of the notation between the kanji notation of a predetermined language and the kanji notation of another language.
[0088]
According to the twelfth aspect of the invention, in the variant notation normalization processing program according to any one of claims 7 to 9, the character correspondence between the kanji notation of a predetermined language and the variant character is described. Various languages with different kanji notations by having both the different notation normalization rules and the different notation normalization rules in which kanji notations of other languages corresponding to the kanji notation of a given language are described It can absorb the fluctuations of various notations.
[0089]
According to the storage medium of the invention of the thirteenth aspect, by storing the different notation normalization processing program according to any one of the seventh to twelfth aspects, the different notation normalization processing program stored in the storage medium is stored. By causing the computer to read it, it is possible to obtain the same operational effects as the invention according to any one of claims 7 to 12.
[0090]
According to the document retrieval device of the invention described in claim 14, the different notation normalization processing device according to any one of claims 1 to 6, which performs the different notation normalization processing, the document input means for receiving the input text, and this When the input text received by the document input means is a document, a document storage means for storing in the document database, and a character string is extracted from the document stored in the document database by the document storage means for each extracted character string After the different notation normalization processing is performed by the different notation normalization processing apparatus, when the input text received by the document input means is an index registration unit that extracts and stores index information, and the search term Search to create a search condition after extracting the character string from the search term and performing the different notation normalization process A matter creation means, a search processing means for searching the index information based on the search condition created by the search condition creation means, and a result output means for outputting a search result by the search processing means, Documents written in Kanji-speaking languages (Chinese (simplified and traditional), Japanese, Korean, Vietnamese, etc.) by generating index information and search conditions that have been subjected to different notation normalization processing Occurrence of a search omission in the document search process for can be prevented, so that the most appropriate document can be searched.
[0091]
According to the document search program of the invention described in claim 15, the different notation normalization processing which is installed in a computer and realizes the different notation normalization processing in the different notation normalization processing device according to any one of claims 1 to 6. A function, a document input function that accepts an input text, a document storage function that stores the document in the document database when the input text received by the document input function is a document, and the document database that is stored in the document database by the document storage function After extracting a character string from a document and performing different notation normalization processing for each extracted character string using the different notation normalization processing function, an index registration function for extracting and storing index information and acceptance by the document input function If the input text is a search word, the search word is sent to the different notation normalization processing function and the search word A search condition creation function for creating a search condition after extracting a character string from the character string and performing a different notation normalization process, and a search for searching the index information based on the search condition created by the search condition creation function A processing function and a result output function for outputting a search result by the search processing function, and generating index information and search conditions subjected to the different notation normalization process, thereby generating Search omissions in the document search process for documents written in languages (Chinese (simplified and traditional), Japanese, Korean, Vietnamese, etc.) can be prevented. can do.
[0092]
According to the storage medium of the invention described in claim 16, by storing the program described in claim 15 and causing the computer to read the document search program stored in the storage medium, Similar effects can be obtained.
[Brief description of the drawings]
FIG. 1 is a block diagram schematically showing a hardware configuration of a different notation normalization processing apparatus according to a first embodiment of this invention.
FIG. 2 is a functional block diagram showing a different notation normalization processing apparatus.
FIG. 3 is an explanatory diagram showing an example of a different notation normalization rule.
FIG. 4 is an explanatory diagram showing an example of a different notation normalization rule.
FIG. 5 is an explanatory diagram illustrating an example of a different notation normalization rule;
FIG. 6 is a flowchart showing an overall flow of different notation normalization processing in the different notation normalization processing apparatus;
FIG. 7 is a functional block diagram illustrating a schematic configuration of a document search apparatus according to a second embodiment of this invention.
FIG. 8 shows a flow of processing in the document search apparatus, (a) is a flowchart showing a flow of index information registration processing for extracting and registering index information from a character string included in a document, and (b) is based on a search word. It is a flowchart which shows the outline of the flow of the document search process which searches a document.
[Explanation of symbols]
1. Different notation normalization processing device
7 Storage media
50 Document retrieval device
52 Document input means
53 Document storage means
54 Document Database
55 Index registration means
56 Index information storage unit
57 Search condition creation means
58 Search processing means
101 Document input means
102a Kanji equivalent cutout means
102b Normalization processing means
102c Different notation normalization rules
102d Different notation normalization rule selection means
102e Normalization rule storage means

Claims (16)

A document input means for receiving input text;
Kanji equivalent cutout means for extracting kanji notation from the input text received by the document input means;
Normalization rule storage means for storing at least two different notation normalization rules describing character correspondence between languages having different kanji notations;
Different notation normalization rule selection means for selecting the different notation normalization rule suitable for the input text received by the document input means;
A different notation normalization processing device comprising: normalization processing means for normalizing the Chinese character notation extracted by the kanji equivalent cutout means according to the different notation normalization rule selected by the different notation normalization rule selection means. The different notation normalization rule selecting means analyzes language information of the input text, and selects the different notation normalization rule based on the analyzed language information.
The different notation normalization processing apparatus according to claim 1. The different notation normalization rule selecting means, when a plurality of the different notation normalization rules are selected based on the analyzed language information, the desired different notation normalization rule from these different notation normalization rules Allow selection of optimization rules,
The different notation normalization processing apparatus according to claim 2. The different notation normalization rule is a description of character correspondence between kanji notation of a predetermined language and its variant characters.
The different notation normalization processing device according to claim 1. The different notation normalization rule is a description of kanji notations of other languages corresponding to kanji notations of a predetermined language.
The different notation normalization processing device according to claim 1. The different notation normalization rule describing the character correspondence between the kanji notation of the predetermined language and its variant character, and the different notation describing the kanji notation of another language corresponding to the kanji notation of the predetermined language Have both normalization rules,
The different notation normalization processing device according to claim 1. Installed on the computer,
Document input function that accepts input text,
A kanji equivalent cutout function for extracting kanji notation from the input text received by the document input function;
A normalization rule storage function for storing at least two kinds of different notation normalization rules describing character correspondence between languages having different kanji notations;
A different notation normalization rule selection function for selecting the different notation normalization rule suitable for the input text received by the document input function;
A normalization processing function for normalizing the kanji notation extracted by the kanji equivalent cutout function according to the different notation normalization rule selected by the different notation normalization rule selection function, and causing the computer to execute the different notation normalization Processing program. The different notation normalization rule selection function analyzes language information of the input text, and selects the different notation normalization rule based on the analyzed language information.
The non-notation normalization processing program according to claim 7. The different notation normalization rule selection function, when a plurality of the different notation normalization rules are selected based on the analyzed language information, the desired different notation normalization rule from these different notation normalization rules Allow selection of optimization rules,
The non-notation normalization processing program according to claim 8. The different notation normalization rule is a description of character correspondence between kanji notation of a predetermined language and its variant characters.
The non-notation normalization processing program according to any one of claims 7 to 9. The different notation normalization rule is a description of kanji notations of other languages corresponding to kanji notations of a predetermined language.
The non-notation normalization processing program according to any one of claims 7 to 9. The different notation normalization rule describing the character correspondence between the kanji notation of the predetermined language and its variant character, and the different notation describing the kanji notation of another language corresponding to the kanji notation of the predetermined language Have both normalization rules,
The non-notation normalization processing program according to any one of claims 7 to 9. A storage medium for storing the different notation normalization processing program according to any one of claims 7 to 12. The different notation normalization processing apparatus according to any one of claims 1 to 6, which performs different notation normalization processing;
A document input means for receiving input text;
If the input text received by the document input means is a document, a document storage means for storing in a document database;
After extracting the character string from the document stored in the document database by the document storage means and performing the different notation normalization processing for each extracted character string in the different notation normalization processing device, the index information is extracted. Index registration means for storing in the index information storage unit;
When the input text received by the document input means is a search word, after sending the search word to the different notation normalization processing device, extracting a character string from the search word, and performing the different notation normalization process , A search condition creating means for creating a search condition,
Search processing means for searching index information in the index information storage unit based on the search conditions created by the search condition creating means;
A result output means for outputting a result of the search by the search processing means;
A document search apparatus comprising: Installed on the computer,
A different notation normalization processing function for realizing different notation normalization processing in the different notation normalization processing device according to any one of claims 1 to 6,
Document input function that accepts input text,
When the input text received by the document input function is a document, a document storage function for storing in a document database;
After extracting a character string from the document stored in the document database by the document storage function and performing the different notation normalization processing for each extracted character string by the different notation normalization processing function, the index information is extracted. An index registration function to be stored in the index information storage unit;
When the input text received by the document input function is a search word, after sending the search word to the different notation normalization processing function, extracting a character string from the search word, and performing the different notation normalization process , Search condition creation function to create search conditions,
A search processing function for searching index information in the index information storage unit based on the search conditions created by the search condition creation function;
A result output function for outputting the search results by this search processing function;
Search program that causes a computer to execute A storage medium for storing the document search program according to claim 15.

Images