JP2010128660A - Text retrieval program, text retrieving device, and text browsing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain an intended retrieval group about text retrieval by combining retrieval conditions with negative conditions. <P>SOLUTION: A text retrieval program is provided to make a computer, which performs text retrieval by combining retrieval conditions that a first character string is included with negative conditions that a second character string is not included, function as an appearance frequency count function part 43 for, when the first character string is included in the second character, counting the appearance frequency of the first and second character strings in the retrieval object text; and function as an appearance frequency comparison/decision function part 44 for comparing the appearance frequency of the first and the second character strings in the retrieval object text, and for, when the appearance frequency of the first and the second character strings is the same, determining that the retrieval object text is not a hit object, and for, when the appearance frequency of the first character string is larger than the appearance frequency of the second character string, determining that the retrieval object text is the hit object, so that an intended retrieval group can be obtained. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a text search program, a text search device, and a text search method, and in particular, performs a text search that combines a search condition that includes a first character string and a negative condition that does not include a second character string. The present invention relates to a program, a text search device, and a text search method.

Conventionally, in order to eliminate extra hits (hereinafter referred to as search noise) in full text search (full text search by N-gram index or full text solid text search: see Patent Document 1, for example) There is a NOT search method that combines a condition that the character string is included (search condition) and a condition that the specific character string is not included (NOT condition or negative condition).
JP 2002-269139 A

  In the conventional NOT search method, in a full-text search, when a character string of a normal search condition (condition word A) is included in a character string of a negative condition (condition word B), a search set as intended is obtained. The problem that it cannot be obtained occurs.

  For example, in the conventional NOT search method, it is desired to search for documents including “log” in the document search, but to exclude documents hit by “program” as search noise. However, “log” and “ When we wanted to hit a document containing both of the program, we couldn't get the search set we intended.

  For example, the character string “program” includes the character string “log”. Therefore, in the conventional NOT search method, since the character string “log” of the normal search condition is included in the character string “program” of the negative condition, a search set as intended cannot be obtained.

  In the conventional NOT search method, a document including the character string “log” of the search condition is searched, and the document including the character string “program” of the negative condition is excluded from the document including “log”.

  Therefore, in the conventional NOT search method, documents including both the search condition character string “log” and the negative condition character string “program” which are not included in the negative condition character string “program” are deleted. As a result, there is a problem that even a document to be hit is excluded from the search set (search result).

  In other words, in the conventional NOT search method, “I want to search for documents including the character string“ log ”, but I want to exclude documents hit by the character string“ program ”as search noise. However, we want to hit documents that contain both “log” and “program” such as “log search program”. There is a problem that a document search such as "cannot be realized no matter what search conditions and negative conditions are given."

  On the other hand, in the NOT search method, search noise can be eliminated by using word breaks (for example, see Patent Document 1) based on morphological analysis. For example, since the word “mobile phone” is divided into “mobile phone” and “phone”, a search with the character string “charging” in the search condition does not hit.

  However, the NOT search method using word breaks by morphological analysis has the following two problems. As a first problem, morphological analysis causes a search omission. Since morphological analysis creates an index only for each word registered in the dictionary, it causes a search omission for a phrase that is not in the index. For example, even if it is desired to hit the word “mobile phone” with the character string “charge” of the search condition, the NOT search method using the word break by morphological analysis is the word “mobile phone” with the character string “charge” of the search condition. Cannot be hit.

  As a second problem, as described above, the NOT search method that uses word breaks by morphological analysis sometimes excludes documents to be hit from the search set. For example, in a NOT search method using word breaks by morphological analysis, “I want to search for documents including the character string“ log ”, but I want to exclude documents that hit the character string“ access log ”as search noise. However, I want to hit other “logs (event log etc.)”. The case is as follows.

  In the NOT search method using word breaks based on morphological analysis, the word break of the character string “access log” is “access / log”. Therefore, if the character string “log” is searched, the character string “access log” is hit. When a document including the character string “access log” is deleted from the search set, for example, a document including both the character strings “event log” and “access log” is also excluded.

  One embodiment of the present invention has been made in view of the above points, and is intended for a text search that combines a search condition that includes a first character string and a negative condition that does not include a second character string. An object of the present invention is to provide a text search program, a text search apparatus, and a text search method capable of obtaining a street search set.

  In order to solve the above problem, an embodiment of the present invention provides a computer that performs a text search that combines a search condition that includes a first character string and a negative condition that does not include a second character string. When the second character string is a character string including the first character string, the number of appearances counts the number of appearances of the first character string and the second character string, respectively, in the search target text. The count function unit compares the number of appearances of the first character string and the second character string in the text to be searched, and the number of appearances of the first character string and the second character string is the same. If the number of occurrences of the first character string is greater than the number of appearances of the second character string, the search target text is determined to be a hit target. Appearance It is a text search program of order to function as the number comparing and determining function unit.

  In addition, what applied the component, the expression, or the arbitrary combinations of the component of one Embodiment of this invention to a method, an apparatus, a system, a computer program, a recording medium, a data structure, etc. is also effective as an aspect of this invention. .

  As described above, according to an embodiment of the present invention, a search set as intended in a text search that combines a search condition that includes the first character string and a negative condition that does not include the second character string. A text search program, a text search device, and a text search method can be provided.

  Next, the best mode for carrying out the present invention will be described based on the following embodiments with reference to the drawings. In this embodiment, in order to distinguish from the conventional NOT search method, the NOT search method of this embodiment is referred to as an extended NOT search method.

  As shown in FIGS. 1 and 2, the extended NOT search method of this embodiment is different from the conventional NOT search method in the search set obtained by the search. FIG. 1 is an image diagram of an example of search processing by a conventional NOT search method. FIG. 2 is an image diagram of an example of search processing by the extended NOT search method of this embodiment.

  The image diagrams of FIGS. 1 and 2 are a document 1 “log search method”, a document 2 “log search program”, and a document 3 “search program” with a search condition character string “log” and a negative condition character string “program”. , Reference 4 “search method” is a NOT search or extended NOT search.

  In the conventional NOT search method shown in FIG. 1, documents 1 to 3 are hit and documents 4 are not hit by searching the character string “log” as a search condition. Subsequently, in the conventional NOT search method, the document 1 is hit and the documents 2 to 3 are not hit by searching the character string “program” as a negative condition.

  In the conventional NOT search method shown in FIG. 1, reference 1 is obtained as a search result. Originally, the “log search program” of reference 2 also needs to be obtained as a search result. This is because the “log search program” in Document 2 includes “log” in the first and second characters in addition to “log” included in the character string “program”.

  Thus, in the conventional NOT search method, when a search condition character string (for example, a log) is included in a negative condition character string (for example, a program), a search set as intended cannot be obtained. .

  In the extended NOT search method of this embodiment shown in FIG. 2, documents 1 to 3 are hit and documents 4 are not hit by searching the character string “log” as a search condition. Subsequently, in the extended NOT search method according to the present embodiment, the documents 1 and 2 are hit and the document 3 is not hit by searching the character string “program” as a negative condition.

  In the extended NOT search method of the present embodiment shown in FIG. 2, documents 1 and 2 as intended are obtained as search results. The extended NOT search method of this embodiment obtains a search set as intended even if a character string (for example, a log) of a search condition is included in a character string (for example, a program) of a negative condition, according to the principle described later. Can do.

  FIG. 3 is a configuration diagram of an example of a system that performs an extended NOT search according to the present embodiment. The system shown in FIG. 3A includes a text search device 10 realized by a stand-alone computer. The system shown in FIG. 3B includes a text search device 20 and a user terminal 21 that are realized by a computer connected to a network 22 such as the Internet or a LAN. The text search device 20 and the user terminal 21 can perform data communication via the network 22.

  The text search apparatus 10 included in the system of FIG. 3A receives search conditions and negative conditions from the user, performs an extended NOT search based on the input search conditions and negative conditions, and then outputs the search results. To do. Also, the text search device 20 included in the system of FIG. 3B accepts the search condition and negative condition input by the user to the user terminal 21 via the network 22 and expands NOT according to the input search condition and negative condition. After performing the search, the search result is provided to the user terminal 21 via the network 22.

  FIG. 4 is a configuration diagram of an example of a text search apparatus. Here, the text search device 20 will be described as an example. 4 includes an input device 31, an output device 32, a drive device 33, an auxiliary storage device 34, a main storage device 35, an arithmetic processing device 36, and an interface device 37 that are connected to each other via a bus B. Configured as follows.

  The input device 31 includes a keyboard and a mouse, and is used for inputting various signals. The output device 32 includes a display device and is used to display various windows, data, and the like. The interface device 37 includes a modem, a LAN card, and the like, and is used to connect to a network 22 such as the Internet or a LAN.

  The text search program for performing the extended NOT search of this embodiment is at least a part of various programs that control the text search device 20. The text search program is provided by distributing the recording medium 38, downloading from the network 22, or the like. The recording medium 38 on which the text search program is recorded is information such as a CD-ROM, flexible disk, magneto-optical disk, etc., a recording medium for recording information optically, electrically or magnetically, ROM, flash memory, etc. Various types of recording media such as a semiconductor memory for electrical recording can be used.

  When the recording medium 38 on which the text search program is recorded is set in the drive device 33, the text search program is installed in the auxiliary storage device 34 from the recording medium 38 via the drive device 33. The text search program downloaded from the network 22 is installed in the auxiliary storage device 34 via the interface device 37.

  The auxiliary storage device 34 stores the installed text search program and also stores necessary files, data, and the like. The main storage device 35 reads and stores the text search program from the auxiliary storage device 34 when the computer is activated. The arithmetic processing unit 36 implements various processes as described later in accordance with a text search program stored in the main storage device 35. The text search program may be software such as an add-on for adding a function to the software in addition to the software main body such as an application and middleware.

  In addition, although the example of the text search device 20 was demonstrated in FIG. 4, the text search device 10 is also the same. However, the text search device 10 may not have the interface device 37.

  Hereinafter, the details of the text search devices 10 and 20 that perform the extended NOT search of this embodiment will be described. It should be noted that the text search devices 10 and 20 are different in whether the search conditions and negative conditions are directly input from the user or the search conditions and negative conditions input by the user to the user terminal 21 are received via the network 22. There is a difference between outputting the result directly or providing the user terminal 21 with the search result and causing the user terminal 21 to output the search result, but the functions are the same. Therefore, in the following description, the text search apparatus 10 will be described as an example.

  FIG. 5 is a block diagram of an example of a text search apparatus that performs an extended NOT search according to this embodiment. The text search apparatus 10 of FIG. 5 includes a condition reception function unit 41, a search function unit 42, an appearance count counting function unit 43, an appearance count comparison / determination function unit 44, a result output function unit 45, and search target text data 46. It is a configuration.

  The processing procedure of the text search apparatus 10 of FIG. 5 will be described with reference to the flowchart of FIG. FIG. 6 is a flowchart illustrating an example of a processing procedure of the text search apparatus that performs the extended NOT search according to this embodiment.

  In the flowchart of FIG. 6, the document search condition is “search for documents including the character string“ log ”. A document including the character string “program” is excluded as search noise. However, documents including the character string “log” in addition to the character string “log” included in the character string “program” are hit like the character string “log search program”. ] Represents the case.

  In step S1, the condition reception function unit 41 receives a search condition and a negative condition from the user. For example, the condition reception function unit 41 receives a character string “log” as a search condition and a character string “program” as a negative condition, such as “log-program”. The condition reception function unit 41 notifies the search function unit 42 of the search conditions and negative conditions input by the user.

  In step S2, the search function unit 42 sets “0” to “n” representing the currently evaluated document. Further, the search function unit 42 sets the number of documents included in the text data 46 to be searched for “m” representing the total number of documents. In step S3, the search function unit 42 adds “1” to “n” representing the currently evaluated document.

  In step S4, the search function unit 42 evaluates whether or not the character string “log” of the search condition and the character string “program” of the negative condition are hit for the first document included in the text data 46.

  In step S 5, the search function unit 42 determines whether or not the first document included in the text data 46 includes the search condition character string “log”. If the first document included in the text data 46 includes the character string “log” of the search condition, the search function unit 42 proceeds to step S 6, and the first document included in the text data 46 has a negative condition. It is determined whether or not the character string “program” is included.

  If the first document included in the text data 46 includes the negative condition character string “program”, the search function unit 42 counts the number of occurrences of the character strings “log” and “program” in the first document. Is requested to the appearance count function unit 43.

  In step S 7, the appearance count counting function unit 43 counts the number of appearances of the character string “log” in the first document based on the request from the search function unit 42. In step S8, the appearance count function unit 43 counts the number of appearances of the character string “program” in the first document based on the request from the search function unit 42. The appearance number counting function unit 43 notifies the appearance number comparison / determination function unit 44 of the counted appearance number of the character string “log” and the appearance number of the character string “program” in the first document.

  In step S9, the appearance number comparison / determination function unit 44 compares the appearance number of the character string “log” with the appearance number of the character string “program” in the counted first document. If the number of appearances of the character string “log” and the number of appearances of the character string “program” in the first document counted are the same, the appearance number comparison / determination function unit 44 proceeds to step S10. The document is determined not to be hit, and the search function unit 42 is notified.

  If the number of occurrences of the character string “log” in the counted first document is the same as the number of appearances of the character string “program”, the first document is included in the negative condition character string “program”. Since the character string “log” is not included in addition to the character string “log”, it is excluded from the hit target.

  If the number of appearances of the character string “log” in the counted first document is larger than the number of appearances of the character string “program”, the appearance number comparison / determination function unit 44 proceeds to step S11 and hits the first document. The target is determined and notified to the search function unit 42. The search function unit 42 adds the first document to the hit document list.

  If the number of appearances of the character string “log” in the counted first document is greater than the number of appearances of the character string “program”, the character string “included in the negative condition character string“ program ”is included in the first document. Since the character string “log” is included in addition to “log”, it is a hit target.

  In step S5, if the character string “log” of the search condition is not included in the first document included in the text data 46, the search function unit 42 proceeds to step S10, and the first document is excluded from the hit target. And the search function unit 42 is notified. In step S6, if the first document included in the text data 46 does not include the negative condition character string “program”, the appearance frequency comparison / determination function unit 44 proceeds to step S11 and proceeds to step S11. Is determined as a hit target, and the search function unit 42 is notified. The search function unit 42 adds the first document to the hit document list.

  Progressing to step S12 following step S10 or S11, the search function unit 42 determines whether “n” representing the currently evaluated document and “m” representing the total number of documents are the same. If “n” representing the currently evaluated document is not the same as “m” representing the total number of documents, the search function unit 42 returns to step S3 and sets “n” representing the currently evaluated document to “n”. 1 ”is added and the process proceeds to the next document. The subsequent processing is the same as in the first document.

  If “n” representing the currently evaluated document and “m” representing the total number of documents are the same, the search function unit 42 determines that all documents included in the text data 46 to be searched have been searched. Then, the process proceeds to step S13. In step S13, the search function unit 42 requests the result output function unit 45 to provide the hit document list to the user. The result output function unit 45 displays, for example, a hit literature list on the display device and provides it to the user.

  As described above, the text search apparatus 10 according to the present embodiment can eliminate search noise that cannot be eliminated by the conventional NOT search by the extended NOT search, regardless of the morphological analysis. In other words, the extended NOT search according to the present embodiment can exclude documents having only character strings to be excluded as search noise.

  The text search apparatus 10 of FIG. 5 is provided with an appearance count function unit 43 and an appearance number comparison / determination function unit 44 separately from the search function unit 42. However, as shown in FIG. 7, the text search device 10 may include an appearance count counting function unit 43 and an appearance count comparison / determination function unit 44 included in the search function unit 42.

  FIG. 7 is a block diagram of another example of a text search apparatus that performs an extended NOT search according to this embodiment. The text search apparatus 10 of FIG. 7 has a configuration including a condition reception function unit 41, a search function unit 42, a result output function unit 45, and text data 46 to be searched. The search function unit 42 includes an appearance count counting function unit 43 and an appearance count comparison / determination function unit 44.

  The processing procedure of the text search apparatus 10 in FIG. 7 will be described with reference to the flowchart in FIG. FIG. 8 is a flowchart of another example showing the processing procedure of the text search apparatus for performing the extended NOT search according to this embodiment.

  The flowchart of FIG. 8 represents a case where the document search conditions are the same as those of the flowchart of FIG. Since the process of step S11-S14 is the same as that of step S1-S4 shown in FIG. 6, description is abbreviate | omitted.

  In step S15, the appearance count counting function unit 43 in the search function unit 42 counts the number of appearances of the character string “log” in the first document. When the number of appearances of the character string “log” is counted one or more times, the appearance number counting function unit 43 proceeds to step S16 and counts the number of appearances of the character string “program” in the first document. When the number of appearances of the character string “program” is counted one or more times, the appearance number counting function unit 43 proceeds to step S17.

  In step S17, the appearance number comparison / determination function unit 44 compares the appearance number of the character string “log” with the appearance number of the character string “program” in the counted first document. If the number of appearances of the character string “log” in the first document and the number of appearances of the character string “program” are the same, the appearance number comparison / determination function unit 44 proceeds to step S18. The document is determined not to be hit.

  If the number of occurrences of the character string “log” in the counted first document is the same as the number of appearances of the character string “program”, the first document is included in the negative condition character string “program”. Since the character string “log” is not included in addition to the character string “log”, it is excluded from the hit target.

  If the number of appearances of the character string “log” in the counted first document is larger than the number of appearances of the character string “program”, the appearance number comparison / determination function unit 44 proceeds to step S19 and hits the first document. Judge as the target. The search function unit 42 adds the first document to the hit document list.

  If the number of appearances of the character string “log” in the counted first document is greater than the number of appearances of the character string “program”, the character string “included in the negative condition character string“ program ”is included in the first document. Since the character string “log” is included in addition to “log”, it is a hit target.

  If the number of appearances of the character string “log” is not counted in step S15 and the number of appearances is 0, the appearance count counting function unit 43 proceeds to step S18 and determines that the first document is not a hit target. To do. If the number of appearances of the character string “program” is not counted in step S16 and the number of appearances is 0, the appearance count counting function unit 43 proceeds to step S19 and determines the first document as a hit target. The search function unit 42 adds the first document to the hit document list.

  Progressing to step S20 following step S18 or S19, the search function unit 42 determines whether or not “n” representing the currently evaluated document and “m” representing the total number of documents are the same. If “n” representing the currently evaluated document and “m” representing the total number of documents are not the same, the search function unit 42 returns to step S13 and sets “n” representing the currently evaluated document to “n”. 1 ”is added and the process proceeds to the next document. The subsequent processing is the same as in the first document.

  If “n” representing the currently evaluated document and “m” representing the total number of documents are the same, the search function unit 42 determines that all documents included in the text data 46 to be searched have been searched. Then, the process proceeds to step S21. In step S21, the search function unit 42 requests the result output function unit 45 to provide the hit document list to the user. The result output function unit 45 displays, for example, a hit literature list on the display device and provides it to the user.

  As described above, the text search apparatus 10 according to the present embodiment can eliminate search noise that cannot be eliminated by the conventional NOT search by the extended NOT search, regardless of the morphological analysis. In other words, the extended NOT search according to the present embodiment can exclude documents having only character strings to be excluded as search noise.

  According to the extended NOT search of the present embodiment, a NOT search is possible even when “a character string of the search condition is included in the character string of the negative condition” that cannot be realized by the way of providing the search condition and the negative condition. Thus, a search result as intended without the search noise can be obtained.

The present invention may have the following configurations as described below.
(Appendix 1)
A computer that performs a text search that combines a search condition that includes the first character string and a negative condition that does not include the second character string,
When the second character string is a character string including the first character string, the number of appearances counts the number of appearances of the first character string and the second character string, respectively, in the search target text. A count function,
The number of appearances of the first character string and the second character string is compared in the search target text, and the search is performed when the number of appearances of the first character string and the second character string is the same. Comparison / determination of the number of appearances when it is determined that the target text is not a hit target, and when the number of appearances of the first character string is greater than the number of appearances of the second character string, the search target text is determined as a hit target A text search program to function as a functional part.
(Appendix 2)
When the first character string and the second character string are included in the search target text, the appearance count counting function unit is configured to display the first character string and the second character string in the search target text. The text search program according to appendix 1, which counts the number of occurrences of each character string.
(Appendix 3)
The text search program according to appendix 1 or 2, wherein the appearance count function unit and the appearance count comparison / determination function unit are provided in a search function unit that performs the text search.
(Appendix 4)
A text search device that performs a text search that combines a search condition that includes a first character string and a negative condition that does not include a second character string,
When the second character string is a character string including the first character string, the number of appearances counts the number of appearances of the first character string and the second character string, respectively, in the search target text. A count function,
The number of appearances of the first character string and the second character string is compared in the search target text, and the search is performed when the number of appearances of the first character string and the second character string is the same. Comparison / determination of the number of appearances when the target text is determined not to be hit and the number of appearances of the first character string is greater than the number of appearances of the second character string A text search apparatus having a function unit.
(Appendix 5)
A text search method for performing a text search that combines a search condition that a computer includes a first character string and a negative condition that the computer does not include a second character string,
The computer counts the number of appearances of the first character string and the second character string in the search target text when the second character string is a character string including the first character string. An appearance count step,
The computer compares the number of appearances of the first character string and the second character string in the text to be searched, and the number of appearances of the first character string and the second character string is the same. The number of appearances when the text to be searched is determined not to be hit and the number of appearances of the first character string is greater than the number of appearances of the second character string A text search method comprising a comparison / determination step.

  The present invention is not limited to the specifically disclosed embodiments, and various modifications and changes can be made without departing from the scope of the claims.

It is an image figure of an example of the search process by the conventional NOT search method. It is an image figure of an example of the search process by the extended NOT search method of a present Example. It is a block diagram of an example of the system which performs the extended NOT search of a present Example. It is a block diagram of an example of a text search device. It is a block diagram of an example of the text search apparatus which performs the extended NOT search of a present Example. It is a flowchart of an example showing the process sequence of the text search apparatus which performs the extended NOT search of a present Example. It is a block diagram of the other example of the text search apparatus which performs the extended NOT search of a present Example. It is a flowchart of the other example showing the process sequence of the text search apparatus which performs the extended NOT search of a present Example.

Explanation of symbols

10, 20 Text search device 21 User terminal 22 Network 31 Input device 32 Output device 33 Drive device 34 Auxiliary storage device 35 Main storage device 36 Arithmetic processing device 37 Interface device 38 Recording medium 41 Condition receiving function portion 42 Search function portion 43 Appearance Number of times count function unit 44 Number of appearances comparison / determination function unit 45 Result output function unit 46 Text data to be searched

Claims (5)

A computer that performs a text search that combines a search condition that includes the first character string and a negative condition that does not include the second character string,
When the second character string is a character string including the first character string, the number of appearances counts the number of appearances of the first character string and the second character string, respectively, in the search target text. A count function,
The number of appearances of the first character string and the second character string is compared in the search target text, and the search is performed when the number of appearances of the first character string and the second character string is the same. Comparison / determination of the number of appearances when it is determined that the target text is not a hit target, and when the number of appearances of the first character string is greater than the number of appearances of the second character string, the search target text is determined as a hit target A text search program to function as a functional part.   When the first character string and the second character string are included in the search target text, the appearance count counting function unit is configured to display the first character string and the second character string in the search target text. The text search program according to claim 1, wherein the number of occurrences of each character string is counted.   The text search program according to claim 1, wherein the appearance count function unit and the appearance number comparison / determination function unit are provided in a search function unit that performs the text search. A text search device that performs a text search that combines a search condition that includes a first character string and a negative condition that does not include a second character string,
When the second character string is a character string including the first character string, the number of appearances counts the number of appearances of the first character string and the second character string, respectively, in the search target text. A count function,
The number of appearances of the first character string and the second character string is compared in the search target text, and the search is performed when the number of appearances of the first character string and the second character string is the same. Comparison / determination of the number of appearances when it is determined that the target text is not a hit target, and when the number of appearances of the first character string is greater than the number of appearances of the second character string, the search target text is determined as a hit target A text search apparatus having a function unit. A text search method for performing a text search that combines a search condition that a computer includes a first character string and a negative condition that the computer does not include a second character string,
The computer counts the number of appearances of the first character string and the second character string in the search target text when the second character string is a character string including the first character string. An appearance count step,
The computer compares the number of appearances of the first character string and the second character string in the text to be searched, and the number of appearances of the first character string and the second character string is the same. The number of occurrences when the text to be searched is determined not to be hit and the number of appearances of the first character string is greater than the number of appearances of the second character string A text search method comprising a comparison / determination step.

Images