JP2005182408A - Term retrieval support system, term retrieval support method and term retrieval support program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a term retrieval support system capable of performing the Kanji (Chinese character) conversion processing of an entered retrieval character string, retrieving a homepage and suggesting a correct term even if part of the retrieval character string is erroneously inputted. <P>SOLUTION: The term retrieval support system is provided with a central processing unit (CPU) 100, a storage device 200, a connection device 300 which acquires a retrieval character string, a program storage device 230, and a data storage device 231. Further, the storage device 200 is provided with a term dictionary 201, a retrieval index table preservation part 202 for preserving index terms for which listed terms included in a file are converted to Kanji, and retrieval object meta data 203. Further, the CPU 100 is provided with a Kanji conversion part 110 for performing the Kanji conversion processing of the entered retrieval character string and outputting a Kanji-converted retrieval character string, an index term retrieval part 112 for retrieving the index term matched with the Kanji-converted character string, an information acquisition part 113 for acquiring the information of the homepage corresponding to the index term, an index preparation part 111, and a similar term conversion part 114. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a term search technology using a computer, and more particularly to a term search support system, a term search support method, and a term search support program that can handle a search character string including kana characters.

An information network represented by the Internet is said to be a treasure trove of information. However, since a conventional search system searches only a homepage file that includes the same character string as the input character string, the Internet user has to select the search character string to be input to the search system accurately. Therefore, in order to reduce the burden of user search string selection, a category reference dictionary in which terms are classified by category is prepared in advance, and even when a homepage that includes a string that matches the input search string is not searched. A search system that extracts a character string in the same category as the input search character string from the category reference dictionary and displays it to the user has been proposed (see, for example, Patent Document 1).
JP 2002-132808 JP

  However, even if a category reference dictionary is prepared in advance, it cannot be handled when the user inputs the search character string itself without using the correct kanji. For example, when children use a conventional search system, children may not be able to enter a search string using the correct kanji because there are unlearned kanji depending on the grade, so characters using only “kana” When a string or a character string that mixes “kanji” and “kana” is entered, the conventional search system can convert the entered search string into terms using the correct kanji and search the home page. There wasn't. Further, even when the user inputs a term using the correct kanji as the search character string, the home page where the term is published in hiragana or the like cannot be retrieved. In some cases, the user may mistakenly enter a part of the characters included in the character string. In such a case, a search system that displays correct terms as correction candidates and continues the search or advises the client is desired. .

  In view of the above-mentioned problems, the present invention performs an input search character string conversion process on the input search character string, searches a file such as a home page in which terms corresponding to the search character string after the Chinese character conversion process are posted, It is an object of the present invention to provide a term search support system, a term search support method, and a term search support program that can advise correct terms even when a part is erroneously input.

  To achieve the above object, the present invention is characterized by (a) a connection device that acquires a search character string, and (b) kanji conversion that performs an input search character string conversion process and outputs a corrected search character string. And (c) a search index table storage unit that stores index terms obtained by performing Kanji conversion processing on the published terms included in the file, and (d) an index term search unit that searches for index terms that match the corrected search character string. (E) The gist of the present invention is a term search support system including an information acquisition unit that acquires information of a file corresponding to an index term.

  The second feature of the present invention is that (a) the connecting device acquires a search character string, and (b) a kanji conversion unit performs kanji conversion processing on the search character string and outputs a corrected search character string; (C) The index creation unit stores the index terms obtained by converting the published terms contained in the file into Kanji, and (d) the index term search unit matches the corrected search character string. The gist of the present invention is a term search support method including a step of searching for an index term to be performed, and (e) an information acquisition unit acquiring a file information corresponding to the index term.

  The third feature of the present invention is: (b) a procedure for obtaining a search character string; (b) a procedure for converting the search character string into a kanji character and outputting a corrected search character string; To obtain index terms that have been converted from Kanji to Kanji conversion terms, (d) to retrieve index terms that match the corrected search string, and (e) to obtain file information corresponding to the index terms This is a term search support program for executing the procedure.

  According to the present invention, an input search character string is subjected to kanji conversion processing, a file containing terms corresponding to the search character string after the kanji conversion processing is searched, and a part of the search character string is erroneously input. In this case, it is possible to provide a term search support system, a term search support method and a term search support program that can advise correct terms.

  Next, embodiments of the present invention will be described with reference to the drawings. In the following description of the drawings, the same or similar parts are denoted by the same or similar reference numerals. The embodiments shown below exemplify apparatuses and methods for embodying the technical idea of the present invention, and the technical idea of the present invention specifies the arrangement of components and the like as follows. Not what you want. The technical idea of the present invention can be variously modified within the scope of the claims.

  As shown in FIG. 1, the term search support system 1 according to the embodiment of the present invention includes a plurality of client computers 10A, 10B, 10C,..., 10N and a plurality of homepage servers 20A, via the Internet 5. 20B, 20C, ..., connected to 20N. Each of the homepage servers 20A to 20N is a computer that stores a homepage file including various information such as a learning purpose and a hobby purpose. Each of the client computers 10A to 10N is a personal computer arranged in an educational institution such as a company or a school or a home.

  Furthermore, the term search support system 1 includes a central processing unit (CPU) 100, a storage device 200, a connection device 300, an input device 302, a program storage device 230, and a data storage device 231 as shown in FIG. The CPU 100 further includes a kanji conversion unit 110, an index term search unit 112, an index creation unit 111, an information acquisition unit 113, a similar term conversion unit 114, and a display instruction unit 116. The storage device 200 includes a term dictionary 201, a search index table. A storage unit 202 and search target metadata 203 are provided.

  Here, as the connection device 300, a router, a terminal adapter (TA), a modem, and the like that can be connected to the Internet 5 shown in FIG. 1 can be used. The connection device 300 transmits / receives information to / from the client computers 10A to 10N and the homepage servers 20A to 20N via the Internet 5.

  The search target metadata 203 shown in FIG. 2 is a list of information included in the homepage file provided by the homepage servers 20A to 20N shown in FIG. In the search target metadata 203, information such as a title, a home page address, a search identification number, an outline, a field, a target age, a copyright, and a use restriction for each of a plurality of home pages is registered. Further, for a homepage having a hierarchical structure, a catalog is stored for each file of each hierarchy.

  The term dictionary 201 is a term database including a list of terms used in each field and a part or all of kanji included in each term converted into hiragana. FIG. 3 shows an example of a term database included in the term dictionary 201. For the registered term “atom”, “genko”, “harashi” Each “Genshi” is registered. In addition, for example, for school education, the terms in the term dictionary 201 converted to hiragana are determined by the Kanji distribution table by school year and the Agency for Cultural Affairs, which are stipulated in the Study Guideline issued by the Ministry of Education, Culture, Sports, Science and Technology. It may be limited to the one converted to hiragana according to the correspondence table for learning lessons learned by school type. Further, the term dictionary 201 may include terms corresponding to each grade according to the kanji distribution table for each grade.

  The kanji conversion unit 110 compares the input character string with the terms registered in the term dictionary 201 and performs kana-kanji conversion processing. For example, if the input string contains only “Kana” such as “Genshi”, or contains both “Kanji” and “Kana” such as “Genko” or “Harashi”, The kanji conversion unit 110 converts the character string input with reference to the term dictionary 201 into kanji characters. If the input character string is already expressed in Chinese characters such as “Atom”, the input character string is maintained. The kanji conversion unit 110 unifies the character types in the input character string. For example, when half-width characters and full-width characters are mixed, the alphabet is unified into half-width characters, and the kana-kanji characters are unified into full-width characters.

  The index creation unit 111 acquires a posted character string included in each of a plurality of homepage files on the Internet registered in the search target metadata 203. Further, the index creating unit 111 divides the acquired posted character string into words by morphological analysis or the like, and extracts unnecessary elements such as symbols from the divided words and defines them as posted terms.

  The index table storage unit for search 202 is stored in the search target metadata 203 for each of the index terms obtained by performing kanji processing on the posted terms extracted by the index creation unit 111 and the files of a plurality of websites corresponding to the index terms. The same search identification number as the search identification number is stored.

  The index term search unit 112 searches the search index table storage unit 202 for an index term that matches the corrected search character string obtained by converting the search character string acquired by the connection device 300 into the kanji conversion unit 110. .

  The information acquisition unit 113 acquires the search identification number of the homepage corresponding to the index term searched by the index term search unit 112 from the search index table storage unit 202, and the corresponding search identification included in the search target metadata 203 Search for a number. Further, the information acquisition unit 113 acquires information on the homepage corresponding to the search identification number from the search target metadata 203.

  The display instruction unit 116 displays the user interface 30 shown in FIG. 4 on the display units of the client computers 10A to 10N connected to the connection device 300 via the Internet 5 shown in FIG. The user interface 30 includes a search character string input area 32, a search button 34, and a search result display area 36. The search character string input area 32 is an area that prompts each user of the client computers 10A to 10N to input a search character string from an input device such as a keyboard. The search button 34 is a button for prompting each user of the client computers 10A to 10N to start searching for a search character string input to the search character string input area 32 from an input device such as a mouse. The search result display area 36 is an area for displaying information acquired by the information acquisition unit 113.

  The similar term conversion unit 114 has an internal counter n, and searches whether a similar term that is different from the input search character string by n characters is registered in the term dictionary 201 according to the value of the internal counter n. When a similar term of n characters is not searched, the value of the internal counter n is incremented by 1. For example, when the value of the internal counter n is 1, the similar term conversion unit 114 searches whether the similar term that is different from the input search character string by one character is registered in the term dictionary 201, and when it is not registered The value of the internal counter n is 2, and the similar term conversion unit 114 searches for similar terms that differ by two characters. In addition, the similar term conversion unit 114 holds variables for the character difference limit value X and the term limit value Z. `` Character difference limit value X '' is the upper limit value of the internal counter n. For example, if 2 is specified for the character difference limit value X, the similar term conversion unit 114 selects a similar term that is two characters different from the search character string. Do the search. The “term limit value Z” is an upper limit value set by the user, the system administrator, etc., for the number of similar terms with n characters that are searched from the term dictionary 201. The similar term conversion unit 114 stops the search when the number of similar terms exceeding the term limit value Z is searched. Each of the character difference limit value X and the term limit value Z is input from the input device 302 or the like.

  The program storage device 230 stores a program for causing the CPU 100 to execute control of data transmission / reception between devices connected to the CPU 100. The data storage device 231 temporarily stores data in the calculation process of the CPU 100. As the input device 302, a keyboard or the like can be used.

  Next, a description will be given of a method of creating the search index table storage unit 202 shown in FIG. 2 used in the term search support method according to the embodiment with reference to FIG.

 (a) In step S401, the index creating unit 111 shown in FIG. 2 operates as one of the homepage servers 20A to 20N shown in FIG. 1 that operates the homepage registered in the search target metadata 203. To access the character string included in the homepage file. Further, the index creating unit 111 converts the full-width character alphabet contained in the acquired posted character string into a half-width character, and the half-width character katakana into a full-width character.

 (b) In step S402, the index creation unit 111 performs a morphological analysis on the acquired posted character string, and divides and extracts the posted term from the posted character string. For example, if the text includes the character string “Tobi box (front forward)”, the text is divided into “top box” and “(front forward)”. Further, in step S403, the index creating unit 111 deletes quotation marks, parentheses, symbols, and the like attached to the terms. For example, in the case of “(forward on table)”, the parentheses are deleted and converted to “forward on table”.

 (c) In step S404, the index creation unit 111 sends the posted terms extracted in steps S402 and S403 to the kanji conversion unit 110, and the kanji conversion unit 110 performs a kanji conversion process on the posted terms with reference to the term dictionary 201. For example, if the posting term is “flying box”, it is converted to “jumping box”. Further, in step S405, the index creating unit 111 stores the search term subjected to the Kanji conversion process as an index term in the search index table storage unit 202 together with the homepage identification number registered in the search target metadata 203. Here, the two index terms “jump box” and “table-forward” are stored in the search index table storage unit 202.

  According to the above method, for example, if “Tobibako Renshu” is described in the homepage file, “jumping box” and “practice” are stored as index terms in the search index table storage unit 202. Further, if there is a description of “the box of jumping box”, “jumping box” and “movement” are stored as index terms in the search index table storage unit 202.

  Next, the term search support method according to the embodiment using the term search support system 1 shown in FIG. 2 will be described using FIG. 6 and FIG.

 (a) In step S601, the display instruction unit 116 illustrated in FIG. 2 is displayed on the display unit connected to the connection device 300 in any of the client computers 10A to 10N illustrated in FIG. Display 30. Here, when a search character string is input to the search character string input area 32 from any of the users of the client computers 10A to 10N and the search button 34 is clicked with a mouse or the like, the connection device 300 is connected via the Internet 5. Get sent search string. Next, in step S602, the connection apparatus 300 gives the acquired search character string to the kanji conversion unit 110, and the kanji conversion unit 110 converts the half-width character alphabet into a half-width character if the search character string includes the full-width character alphabet. If it contains a katakana character, it is converted to a katakana character.

 (b) Further, in step S603, the kanji conversion unit 110 performs kanji conversion processing on the search character string with reference to the term dictionary 201, and outputs a corrected search character string. For example, when “Tobibako” or “Tobi Box” is input as a search character string, the Kanji conversion unit 110 converts Kanji into “Jump Box” by referring to the database shown in FIG. And output as a corrected search character string. If the same term as the search character string is not registered in the term dictionary 201, the kanji conversion unit newly defines the search character string as a corrected search character string without conversion. The kanji conversion unit 110 outputs the corrected search character string to the index term search unit 112.

 (c) In step S604, the index term search unit 112 accesses the search index table storage unit 202, and in step S605, the index term search unit 112 in the search index table storage unit 202 matches the corrected search character string. Search for a term. In step S606, the index term search unit 112 determines whether or not an index term that matches the corrected search character string exists in the search index table storage unit 202. If it exists, the information acquisition unit 113 stores the index term search unit. An identification number corresponding to the index term registered in 202 is given, and the process proceeds to step S607.

 (d) In step S607, the information acquisition unit 113 accesses the search target metadata 203, queries the extracted homepage identification number, and provides information such as the homepage address, summary, target user, copyright, and usage restrictions. The acquired information is output to the display instruction unit 116. Next, the display instruction unit 116 transmits the search character string via the connection device 300 so as to display the homepage information acquired in any of the client computers 10A to 10N shown in FIG. 1 on a display unit such as a monitor. Give instructions. In the search result display area 36 of the user interface 30 displayed on one of the display units of the client computers 10A to 10N, as shown in FIG. 8, the search target metadata for the home page searched by the index term search unit 112 is displayed. Information such as the title, outline, and copyright registered in 203 is displayed, and the term search support method ends.

 (e) If the index term search unit 112 determines that the term matching the corrected search character string does not exist in the search index table storage unit 202 in step S606, the index term search unit 112 uses the corrected search character string as a similar term. The process proceeds to step S701 in FIG. 7 after being given to the conversion unit 114, and the similar term conversion unit 114 defines the internal counter n as 1.

 (f) In step S702, the similar term conversion unit 114 searches the term dictionary 201 for a similar term that differs from the search character string input in step S601 by n characters. For example, if the internal counter n is 1 and “jumping box” has been input as a search character string in step S601, the term “jumping box” using the correct kanji character by one character is searched. In step S703, the similar term conversion unit 114 determines whether or not a similar term having n characters differs in the term dictionary 201. If a similar term having n characters is present in the term dictionary 201, the process proceeds to step S704. If not, the process proceeds to step S707.

 (g) In step S704, the similar term conversion unit 114 compares the search character string searched from the term dictionary 201 with the number of similar terms different in n characters and the set term limit value Z. For example, when the term limit value is set to 5, if similar terms different in n characters are searched for 5 or less, the process proceeds to step S705, and if 6 or more are searched, the process proceeds to step S706. In step S705, the index term search unit 112 lists the similar terms that have been searched for n characters and supplies them to the display instruction unit 116. As shown in FIG. 9, the display instruction unit 116 displays a similar term with n character differences in the search result display area 36 of the user interface 30 displayed on one of the display units of the client computers 10A to 10N shown in FIG. Is displayed and the term search support method is terminated.

 (h) In step S707, the similar term conversion unit 114 adds 1 to the internal counter n to define a new internal counter n. Further, in step S708, the index term search unit 112 determines whether or not the value of the internal counter n is larger than the character difference limit value X. If the value of the internal counter n is equal to or less than the character difference limit value X, the process returns to step S702 and the search for similar terms with n character differences is repeated. If the internal counter n is greater than the character difference limit value X, the process proceeds to step S706. In step S706, the display instruction unit 116 displays on the display unit of any of the client computers 10A to 10N shown in FIG. 1 that the home page corresponding to the search character string input in step S601 has not been searched, and the term End the search support method.

  As described above, according to the term search support system 1 shown in FIG. 2 and the term search support method shown in FIGS. 6 and 7 according to the embodiment, a character string using only “Kana” as a search character string, Even when a character string in which “kanji” and “kana” are mixed is input, the kanji conversion unit 110 shown in FIG. 2 performs kanji conversion processing on the search character string by referring to the term dictionary 201, and the method shown in FIG. By referring to the search index table storage unit 202 created in the above, it is possible to search a home page on which terms using correct kanji corresponding to the search character string are posted. In addition, even when terms using correct kanji are input as a search character string and the terms are expressed only on the homepage as “Kana”, by referring to the search index table storage unit 202 shown in FIG. It is also possible to search for websites that contain only the term “kana”. Therefore, even when a child or the like enters a search string of kanji or kana, it is possible to search a home page that includes terms expressed only in kana, and parents or guardians such as children can search using accurate terms. Even when a character string is entered, it is possible to search a home page on which terms are written in kana.

  Furthermore, even if a part of the search character string contains an incorrect Chinese character, the similar term conversion unit 114 refers to the term dictionary 201 to extract the similar term and presents it to the user, so the user uses the correct Chinese character. The home page search can be resumed by the search character string. Further, when many similar terms are extracted, it may be more appropriate to display that the user has no search result. On the other hand, according to the term search support method shown in FIG. 7, it is possible to determine the upper limit of the number of similar terms shown to the user by comparing the number of similar terms with the term limit value Z in step S704. It becomes.

  In addition, if terms included in the term dictionary 201 are limited to terms that use kanji in the grade-specific kanji distribution table, it is possible to perform term search more quickly and accurately in use in school education.

  The term search support method described above can be expressed as a series of processes or operations connected in time series. Therefore, in order to execute the term search support method in the computer system, the term search support method shown in FIGS. 5, 6, and 7 is realized by a computer program product that specifies a plurality of functions performed by a processor in the computer system. Is possible. Here, the computer program product refers to a recording device or a recording medium that can be input and output to a computer system such as the program storage device 230 shown in FIG. The recording medium includes a memory device, a magnetic disk device, an optical disk device, and other devices capable of recording other programs.

(Other embodiments)
As described above, the present invention has been described according to the embodiment. However, it should not be understood that the description and drawings constituting a part of this disclosure limit the present invention. From this disclosure, various alternative embodiments, examples and operational techniques should be apparent to those skilled in the art. For example, if the terms to be registered in the term dictionary 201 shown in FIG. 2 are limited to terms corresponding to the field, and the home page to be registered in the search target metadata 203 is also limited to a specific field, a home page for each field can be created at high speed. It becomes possible to search efficiently. In the embodiment, the case where a homepage on the Internet is searched is shown. However, the present invention can also be applied to a file search on an information network such as a LAN or a file search in a hard disk of a stand-alone computer. Of course. Thus, it should be understood that the present invention includes various embodiments and the like not described herein. Therefore, the present invention is limited only by the invention specifying matters in the scope of claims reasonable from this disclosure.

1 is a schematic diagram of the Internet to which a term search support system according to an embodiment of the present invention is connected. It is a block diagram which shows the term search assistance system which concerns on embodiment of this invention. It is an example of the table | surface contained in the term dictionary which concerns on embodiment of this invention. It is an example (the 1) of the display screen displayed on the computer connected to the term search assistance system which concerns on embodiment of this invention. FIG. 4 is a flowchart (part 1) illustrating a term search support method according to an embodiment of the present invention. FIG. 5 is a flowchart (part 2) illustrating the term search support method according to the embodiment of the present invention. FIG. 9 is a flowchart (part 3) illustrating the term search support method according to the embodiment of the present invention. It is an example (the 2) of the display screen displayed on the computer connected to the vocabulary search assistance system which concerns on embodiment of this invention. It is an example (the 3) of the display screen displayed on the computer connected to the vocabulary search assistance system which concerns on embodiment of this invention.

Explanation of symbols

1 ... Term search support system
5 ... Internet
10A, 10B, 10C, ..., 10N ... Client computer
20A, 20B, 20C, ..., 20N ... Homepage server
30… User interface
32 ... Search string input area
34 ... Search button
36 ... Search result display area
100 ... Central processing unit (CPU)
110… Kanji conversion part
111 ... Index creation section
112 ... Index term search part
113 ... Information acquisition unit
114… Similar term conversion part
116 ... Display instruction section
200 ... Storage device
201 ... Term dictionary
202… Search index table storage
203 ... Search target metadata
230 ... Program storage device
231 ... Data storage device
300 ... Connecting device
302 ... Input device

Claims (7)

A connected device that retrieves the search string;
A kanji conversion unit that performs a kanji conversion process on the input search character string and outputs a corrected search character string;
An index table storage unit for search that stores index terms obtained by converting Kanji conversion of published terms contained in the file;
An index term search unit that searches for the index terms that match the corrected search character string;
A term search support system comprising: an information acquisition unit that acquires information of the file corresponding to the index term.
The term search support system according to claim 1, further comprising a similar term conversion unit that searches for similar terms similar to the search character string.
The term search support system according to claim 1, further comprising an index creation unit that performs a morphological analysis on a posted character string included in the file and extracts the posted term.
The connected device obtains a search string;
A kanji conversion unit performing kanji conversion processing on the search character string and outputting a corrected search character string;
An index creating unit storing an index term obtained by converting the published term contained in the file into a kanji character in a search index table storing unit;
An index term search unit searching for the index term that matches the corrected search character string;
An information acquisition unit includes the step of acquiring information of the file corresponding to the index term.
The term search support method according to claim 4, further comprising a step of searching for a similar term similar to the search character string by the similar term conversion unit.
6. The step of storing in the search index table storage unit includes a step in which an index creation unit performs morphological analysis on a posted character string included in the file and extracts the posted term. The term search support method described.
On the computer,
Get the search string,
A procedure for converting the search character string into kanji and outputting a corrected search character string;
To obtain index terms that have been converted from Kanji to the published terms contained in the file,
Searching the index terms that match the corrected search string;
A term search support program for executing the procedure of acquiring information of the file corresponding to the index term.

Images