JPH1145250A - Information retrieval device, generating method for retrieval condition using retrieval result, and computer-readable recording medium where program making computer implement same method is recorded - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the usability of information retrieval by allowing a retrieving means to performs a retrieving process according to a 2nd retrieval condition set by a setting means. SOLUTION: This device is equipped with clients 100 which output retrieval conditions for retrieving desired documents and a retrieval engine which can generate a dislocation file 102 from a document data base(DB) 101 and perform vector space retrieval and Boolean retrieval and also equipped with a retrieval server 103 which inputs the retrieval conditions from the clients 100 and retrieves the corresponding documents by using the dislocation file 102 according to the inputted retrieval conditions. Then a 1st retrieval condition is set and a retrieving process is performed according to the 1st retrieval condition; and at least one of pieces of information obtained from the retrieval result is set as a restriction condition and used to set a 2nd retrieval condition. Further, the retrieval server 103 performs a retrieving process according to the 2nd set retrieval condition.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention makes it possible to efficiently set search conditions for the next search process by using a previous search result when a search process must be performed a plurality of times. As described above, the present invention relates to an information search apparatus that improves the convenience of information search, a method of generating search conditions using search results, and a computer-readable recording medium storing a program for causing a computer to execute the method.

[0002]

2. Description of the Related Art With the development of computers, documents recorded and stored using paper as a medium have been developed.
It has become common to handle it as electronic information,
A large number of electronic documents are to be stored in a database.

[0003] By the way, as a large number of digitized documents are accumulated, a search technology that can easily find a desired document from a large number of documents becomes important. In particular, in today's world where everyone works on a computer, it is possible to construct a search system that allows even a novice computer user to easily search for a desired document without using a special technique. This is an important point.

[0004] As one technique for facilitating retrieval, a query (Quu) using a natural language usually used by humans is used.
ery) can be input. This technology has been adopted in many search systems with the development of natural language processing technology in recent years.

[0005]

However, as described above, even if a search system capable of inputting a query in a natural language is constructed, it is difficult to obtain satisfactory search results in one search process. In order to obtain a desired search result, there is a problem that it is necessary to refer to the already performed search result, input a query again, and execute a search process. In other words, using the previous search result, a query suitable for obtaining a desired search result has to be considered and the considered query must be input.

[0006] Further, depending on the purpose of the search, there is a case where a desired search result cannot be obtained unless a plurality of search processes are performed. There is a problem that it is necessary to refer to, input a query again, and execute a search process.

The present invention has been made in view of the above, and when search processing must be performed a plurality of times, the search conditions for the next search processing can be efficiently performed by using the previous search result. So that it can be set well,
An object is to improve the convenience of information retrieval.

[0008]

In order to achieve the above object, an information retrieval apparatus according to claim 1 sets a first retrieval condition in an information retrieval apparatus which performs retrieval processing based on a designated retrieval condition. First setting means; searching means for performing a search process based on the first search condition set by the first setting means; at least one of information obtained from a search result based on the first search condition; Designating means for designating one as a constraint condition, and setting means for setting a second search condition using the constraint condition designated by the designating means,
The search means performs a search process based on the second search condition set by the setting means.

According to a second aspect of the present invention, there is provided an information search device for performing a search process based on a specified search condition, wherein the first setting means sets a first search condition;
A search unit for performing a search process based on the first search condition set by the first setting unit; and at least one of information obtained from a search result based on the first search condition is specified as a constraint condition. And designating means for additionally designating a new search condition, and setting means for setting a second search condition using the constraint condition and the search condition designated by the designating means, wherein the search means comprises: The search processing is performed based on the second search condition set by the setting means.

According to a third aspect of the present invention, there is provided a method for generating a search condition using a search result, comprising: performing a search process based on a set first search condition; And a setting step of setting a second search condition using the constraint condition specified in the specifying step.

According to a fourth aspect of the present invention, there is provided a method for generating a search condition using a search result, comprising: performing a search process based on a set first search condition; A specifying step of specifying at least one of the specified information as a constraint condition and additionally specifying a new search condition, and setting a second search condition using the constraint condition and the search condition specified in the specifying step Setting step to be performed.

A computer-readable recording medium according to a fifth aspect of the present invention has recorded thereon a program for causing a computer to execute each step of the search condition generating method using the search result according to the third or fourth aspect. Things.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An information retrieval apparatus, a method for generating a retrieval condition using a retrieval result, and a computer-readable recording medium on which a program for causing a computer to execute the method will be described below.
This will be described in detail with reference to the accompanying drawings.

FIG. 1 is a system configuration diagram of an information retrieval apparatus according to the present embodiment. The information search device shown in FIG. 1 generates a plurality of clients 100 that output search conditions for searching for a desired document, an inverted file 102 from a document DB (database) 101, and performs a vector space search and a Boolean search. Search engines that can be used (for example, CLARIT of CLARITECH)
), And input search conditions from the client 100. Based on the input search conditions, the transposed file 102
A search server 103 for searching for a corresponding document by using
And a network 104 for connecting the client 100, the search server 103, and the like.

In FIG. 1, a document DB 101 stores a plurality of documents, for example, documents such as newspaper articles. In the present embodiment, a newspaper article is taken as an example of a document stored in the document DB 101, and this newspaper article is a search target. However, the search target is newspaper articles and document DB1
However, the present invention is not limited to the document in FIG.

The transposed file 102 defines the relationship between a plurality of documents in the document DB 101 and a plurality of index terms extracted from each of the plurality of documents by a method described later. An important vocabulary is indicated using a vector expression, and a corresponding document can be searched using the index word.

More specifically, one document is divided into sub-documents consisting of a plurality of sentences in advance, and the noun phrases serving as the index words are extracted from the sub-documents. Statistical information such as an appearance frequency and distribution in the entire document DB 101 is obtained, and each sub-document is converted into a vector expression using the obtained statistical information for each noun phrase. Then, a vector expression of the document is generated based on the vector expression of the converted sub-document. The transposition file 102 stores the document in the document DB 101 thus vectorized.

Each index word can be weighted according to the degree of importance in the corresponding document. Also, the vector expression of the document may be generated based on the vector expression of the sub-document when performing an actual search.

Client 100 and search server 10
Reference numeral 3 includes a personal computer, a workstation, and the like. FIG. 2 is a schematic block diagram showing the processing of the client 100. Client 100
Is a query written in natural language (Quer
y) The user inputs 200 and performs analysis processing such as morphological analysis and syntax analysis using the dictionary 202 storing part of speech information and the like and the grammar dictionary 203 for analyzing the dependency of each word. , Based on the result of the analysis processing, the query
4 and a search term 205 consisting of a noun phrase corresponding to the index term of the transposed file 102 are extracted, and the separated constraints 204 and search terms 205
6 is provided with a natural language processing module 201 for outputting to the search server 103.

The client 100 inputs a search result by the search server 103, displays the input search result, and displays attribute information of a document included in the displayed search result, for example, a creation date of the document, a creator, and the like. The attribute information specified via the specifying unit 208 for specifying the document information of the document is extracted as a constraint, and the natural language processing module 2
01 is provided. As will be described in detail later, the constraint condition setting unit 207 and the specifying unit 208 are used when performing a search process again using the previous search result.

Note that the constraints 204 do not relate to items described in the text of the document, but to document information such as the date of creation of the document and the creator. On the other hand, search term 2
Reference numeral 05 relates to matters described in the text of the document. That is, a document corresponding to the search term 205 is searched, and further, based on the constraint condition 204, the search term 205 is searched.
A process for narrowing down the search results with reference to the document information such as the creation date and the creator of the document corresponding to. Here, narrowing down using the constraint conditions 204 is performed on the document information, but narrowing down may be performed on dates in the text.

FIG. 3 is a schematic block diagram showing the processing of the search server 103. The search server 103 performs a process of registering a document in the document DB 101 in the transposed file 102 and a search process using a vector space search and a Boolean search.

This search server 103 stores the document DB 101
Input the document from, and recognize the format of the document,
Analysis processing such as morphological analysis, syntax analysis, and noun phrase extraction is performed using a dictionary 301 storing part-of-speech information and a grammar dictionary 302 for analyzing the dependency of each word, and the like. Natural language processing module 3 for generating a document set 303 including a list
00, the document set 303 generated by the natural language processing module 300 is input, each sub-document in the input document set 303 is converted into a vector expression, and the vector of the document is converted based on the vector expression of the sub-document. A database build component 304 for generating an expression and registering it in the transposition file 102;
And a search condition 206 consisting of a search term 205 and
For each noun phrase constituting the search term 205 of the input search condition 206, the appearance frequency in the query 200, the document DB
A query build component 305 that obtains statistical information such as distribution in the entire 101 and generates a query document 306 in which the search term 205 is converted into a vector expression using the obtained statistical information for each noun phrase;
The query document 306 generated by the build component 305 is input, a vector of each document in the transposed file 102 is compared with a vector of the query document 306, and a score according to the similarity with the query document 306 is calculated. After a list of documents assigned to each document and having a score exceeding a predetermined threshold is generated, and a process of further narrowing down the documents in the list using the constraint condition 204 is performed, the document list 30 is obtained as a search result.
And a search engine 307 that outputs the search engine 8.

Although FIG. 1 shows a configuration in which the document DB 101 and the transposed file 102 are independently connected to the network 104, a configuration in which these are directly connected to the search server 103 may be used. In FIG. 1, the information retrieval apparatus according to the present embodiment is connected to a network 104.
However, the configuration is such that the processing of the client 100 shown in FIG. 2 is performed by the natural language processing module 300 of the search server 103 shown in FIG. The information search device of the present embodiment can be configured by a computer.

Next, the operation of the information retrieval apparatus having the above-described configuration will be described with respect to (1) an inverted file generation process,
(2) Details will be described in the order of newspaper article search processing.

(1) Transposition File Generation Process FIG. 4 is a flowchart showing a transposition file generation process. Each time a new document, that is, a newspaper article is registered in the document DB 101 (S401), the search server 103 starts processing for inputting the newspaper article and registering it in the transposed file 102 (S402).

In the search server 103, the natural language processing module 300 performs processing for analyzing the newspaper article input in step S402 (S403). In particular,
Processing is performed to determine the format of the input newspaper article, such as a word processing document or a structured document such as HTML. After that, the dictionary 301 and the grammar dictionary 30
2 to perform a morphological analysis, a syntax analysis such as dependency, etc., to divide a newspaper article into a plurality of sub-documents, and to perform processing such as extracting a noun phrase from the divided sub-documents.

Then, the natural language processing module 300
Is based on the result of the processing in step S403.
A noun phrase list is generated for each sub-document, and a document set 303 including the generated noun phrase list is generated (S404).

After that, the database build component 304 inputs the document set 303 generated by the natural language processing module 300, and performs a process of registering the vector expression of the document in the transposed file 102 (S405).

Specifically, the document set 303
Transpose file 102 for each noun phrase of subdocument in
The statistical information such as the frequency of occurrence in the sub-document and the distribution in the entire document DB 101 is obtained as an index term, and the sub-document is converted into a vector expression using the obtained statistical information for each noun phrase. This processing is performed for all the sub-documents in the document set 303, and based on the converted sub-document vector expression, a vector expression of the document is generated and registered in the transposed file 102.

(2) Newspaper Article Retrieval Processing Next, newspaper article retrieval processing will be described. Here, in order to investigate what the Bank of Japan's remarks are affecting the yen market, we searched for newspaper articles that reported that the yen had fluctuated. Searching for articles on the impact on the Japanese yen market (normal search processing), searching for articles on the BOJ's statement (using search results) Search processing: corresponds to a method of generating search conditions using search results of the present invention).

Searching for Articles Related to Yen Market Price FIG. 5 is a flowchart showing a normal search process.
Natural language processing module 201 of client 100
Inputs the query 200 input by the user (S501). Here, as query 200,
Enter "Impact on the yen market".

The natural language processing module 201 analyzes the input query 200 (S502). Specifically, syntax analysis processing such as morphological analysis and dependency is performed using the dictionary 202 and the grammar dictionary 203.

Subsequently, the natural language processing module 201
Extracts the constraint condition 204 and the search term 205 from the query 200 based on the analysis result in step S502, and sets the search condition 206 (S503). The “impact on the yen market”, which is an example of the query 200, does not include the constraint condition 204, and here the “impact on the yen market” or “the yen market” and “impact” are extracted as search terms 205. Is done.

Restriction condition 2 set as described above
The search condition 206 including the search term 04 and the search term 205 is output from the client 100 to the search server 103.

Then, the query build component 305 of the search server 103 inputs the search condition 206 including the constraint condition 204 and the search word 205 from the client 100, and searches the search word 20 of the input search condition 206.
For each of the noun phrases constituting No. 5, statistical information such as the frequency of appearance in the query 200 and the distribution in the entire document DB 101 is determined, and the search term 20 is calculated using the statistical information for each noun phrase.
Query document 3 which converted 5 into vector expression
06 is generated (S504).

The search engine 307 inputs the query document 306 generated by the query build component 305, compares the vector of each newspaper article in the transposed file 102 with the vector of the query document 306, and A score corresponding to the degree of similarity to the document 306 is given to each newspaper article, and a list of newspaper articles given a score exceeding a predetermined threshold is generated (S505). That is, a search process based on a vector space search is performed, and the document DB 10
Generate a list of newspaper articles in 1.

The score according to the similarity expresses the similarity between each newspaper article and the query document 306 on the basis of the cosine distance. A newspaper article having a higher score is more similar to the query document 306. It shows that you are doing. A score threshold is set in the search engine 307 in advance, and the document list 30 including newspaper articles to which a score exceeding the threshold is assigned.
8 is generated as a search result.

When the constraint condition 204 of, for example, one year's newspaper article is extracted from the query 200, the process of narrowing down to one year's newspaper article is also performed in step S505.

Thereafter, the search engine 307 outputs the generated document list 308 to the client 100 as a final search result, and the document list 308 is displayed on the screen of the client 100 (S506). The user of the client 100 selects an arbitrary newspaper article from the document list 308 displayed on the screen, so that the document DB 10
1 can be displayed on the screen.

Search for Articles Related to the Statement by the Governor of the Bank of Japan (Search Processing Using Search Results) Next, search processing using the results of searching for articles regarding the influence on the yen market will be described. FIG. 6 is a flowchart illustrating a search process using a search result.

When a newspaper article, such as the influence of the yen exchange rate, is searched, a document list 308 displayed as a result of the search displays the article headline, newspaper company name, date, and the like as attribute information. Therefore, in order to perform a process of searching for an article with a statement made by the Governor of the Bank of Japan with one week before the date of publication of the article as a constraint, the user must input the date (Designation of attribute information: S601).

More specifically, for example, when five articles that convey the fluctuation of the yen market are displayed as search results, and it is desired to know the effect of the Governor's statement on all five articles, , Specify the dates of all articles. If you want to know the effect of the Governor's remarks on some articles, specify the date of the article you want to know. Further, here, the date of the specified article is "1".
Specify a period such as "week ago", "one month ago", "one year ago", and so on. Here, "one week ago" is specified.

In step S601, attribute information such as a date and an icon for designating a period may be displayed on the screen. A period such as “one week ago” may be input at the same time as the input of the query 200 in step S603 described below.

When the article date (attribute information) is specified in step S601, the constraint condition setting unit 207 extracts the specified date as a search constraint condition and outputs it to the natural language processing module 201 (S602). .

Then, after designating the date and the period in step S602, the user inputs the query 200 (S603). Here, for example, “the statement of the governor of the Bank of Japan” is input as the query 200.

Note that either of the attribute information designation in step S601 and the query input in step S603 may be performed first. When the search results are simply narrowed down by the attribute information (date or the like) obtained from the previous search results, there is no need to input the query 200 in step S603. Therefore, in this case, execution of steps S606 and S607 described later is omitted.

Thereafter, the natural language processing module 201
Inputs the designation of the date and the period from the constraint condition setting unit 207, inputs the query 200, and performs a process of analyzing them (S604). Specifically, the dictionary 20
2 and a grammar dictionary 203 to perform syntax analysis such as morphological analysis and dependency.

Subsequently, the natural language processing module 201
Sets the search condition 206 based on the analysis result in step S502 (S605). In particular,
Using the specification of the date and period input from the constraint condition setting unit 207, “date range = YYMMDD to YYMMD
D "is set. Also, based on the input query 200, the constraint conditions 204 and the search term 20
5 is extracted. The “statement of the Governor of the Bank of Japan”, which is an example of the query 200, does not include the constraint condition 204. Here, the “statement of the Governor of the Bank of Japan” or “the Governor of the Bank of Japan” and “utterance” are set as the search terms 205. . This step S
By the processing in 605, the constraint condition 204 and the search term 2
05 is set.

The constraints 204 set in this way
The search condition 206 including the search term 205 and the search term 205 is output from the client 100 to the search server 103.

Then, the query build component 305 of the search server 103 inputs the search condition 206 including the constraint condition 204 and the search word 205 from the client 100, and executes the query / build as described in step S504 of FIG. The document 306 is generated (S606).

The search engine 307 inputs the query document 306 generated by the query build component 305, compares the vector of each newspaper article in the transposed file 102 with the vector of the query document 306, and A score corresponding to the degree of similarity with 306 is assigned to each newspaper article, and a list of newspaper articles to which a score exceeding a predetermined threshold has been assigned, that is, articles relating to statements by the Governor of the Bank of Japan is generated (S607). That is, a search process is performed to generate a list of newspaper articles in the document DB 101 corresponding to the search term 205.

The score according to the similarity expresses the similarity between each newspaper article and the query document 306 based on the cosine distance. A newspaper article having a higher score is more similar to the query document 306. It shows that you are doing. Then, a score threshold is set in the search engine 307 in advance, and newspaper articles to which a score exceeding the threshold is given are set as search result candidates.

Subsequently, the search engine 307 refers to the attribute information (document information) of each newspaper article, narrows down the search results using the constraint conditions 204, and obtains the document list 308.
Is generated (S608). Specifically, step S60
Referring to the attribute information of the articles in the list created in step 7,
A newspaper article corresponding to the constraint condition 204 of “date range = YYMMDD to YYMMDD” is selected (Boolean search).

In the present embodiment, the order is such that the search processing based on the search term 205 is performed in step S607, and then the refinement is performed in accordance with the constraint condition 204 in step S608, but the order may be reversed. Similar search results can be obtained.

After that, the search engine 307 outputs the generated document list 308 to the client 100 as a final search result, and the document list 308 is displayed on the screen of the client 100 (S609). By selecting an arbitrary article from the document list 308 displayed on the screen, the user of the client 100 can check what statement by the Governor of the Bank of Japan has affected the yen market.

In the above description, an example of searching for articles relating to fluctuations in the yen market has been described. However, the information search apparatus according to the present embodiment can easily perform the following search processing.

First, in a newsgroup or the like,
This is a process of searching for articles related to "automobile" and searching for articles unrelated to automobiles using the poster of the relevant article as a constraint. By this search process, it is possible to investigate what other topics people who are interested in cars are interested in, and to use the information search device of the present embodiment for marketing and the like.

Second, a patent publication relating to “machine translation” is searched to obtain a company name described in the relevant publication, and “document search” is performed using the obtained company name as a constraint.
Searching for patent publications related to "Machine Translation" and "Document Search" allows you to search for companies that have applied for both.

As described above, according to the information retrieval apparatus of the present embodiment, when the retrieval processing must be performed a plurality of times, the attribute information such as the document indicated as the previous retrieval result is restricted. Since the search condition is automatically set together with the newly input query, the search condition for the next search process can be set efficiently, and the convenience of information search can be improved. . In addition, since natural language can be used for the query 200, and the search processing using the vector space method and the Boolean search based on the constraint conditions 204 are used together,
It is possible to easily perform complicated search processing that is difficult only by the search using the vector space method or the Boolean search.

In the above explanation, what kind of statement by the Governor of the Bank of Japan has influenced the yen market
It has been explained that the search process is performed in two steps, that is, the search for the article on "Effect of the yen market" and the search for the article on "BoJ's statement". However, the information search device of the present embodiment can generate the search condition 206 including the constraint condition 204 and the search term 205 from the query 200, By inputting the statement of the Governor of the Bank of Japan, the above-described search processing can be performed in a single step.

In the document list 308 which is a search result obtained by the information search apparatus according to the above-described embodiment, for a document which the user considers appropriate or unsuitable as a search result, the result is stored in a search server. 1
03 can be fed back. That is, the user can specify a positive weight, for example, “+” for a document that is considered to be appropriate as a search result, and a negative weight, for example, “−” for a document that is not appropriate for the search result. Can be specified. as a result,
When the input weight is a positive specification, the weight of the corresponding document in the transposed file 102 is strengthened, and when the input weight is a negative specification, the weight of the document is weakened.

Further, the information search apparatus and the method of generating search conditions using search results described in the present embodiment are as follows:
It is realized by executing a prepared program on a computer or a workstation. This program is for hard disk, floppy disk, CD
-The program is recorded on a computer-readable recording medium such as a ROM, an MO, and a DVD, and is executed by being read from the recording medium by the computer. This program can be distributed via the recording medium or via a network.

[0064]

As described above, according to the information search apparatus of the present invention (claim 1), the first setting means for setting the first search condition and the first setting means set the first search condition. Searching means for performing a search process based on the first search condition, specifying means for specifying at least one of information obtained from a search result based on the first search condition as a constraint condition, and specifying by the specifying means Setting means for setting a second search condition using the set constraint condition, wherein the search means performs a search process based on the second search condition set by the setting means, a plurality of times. When it is necessary to perform search processing over a long time, it is possible to efficiently set search conditions for the next search processing by using the previous search result, and to improve the convenience of information search. .

Further, the information retrieval apparatus of the present invention (Claim 2)
According to the first aspect, the first setting means for setting a first search condition, the search means for performing a search process based on the first search condition set by the first setting means, and the first search Specifying means for designating at least one of the information obtained from the search results based on the condition as a constraint condition, and additionally designating a new search condition; and using the constraint condition and the search condition designated by the designating means, Setting means for setting a second search condition, wherein the search means performs a search process based on the second search condition set by the setting means,
When multiple search processes need to be performed, the previous search results can be set efficiently using new search conditions, and the search conditions for the next search process can be set efficiently. Improvement can be achieved.

Further, according to the search condition generating method using the search result of the present invention (claim 3), after performing the search processing based on the set first search condition, the first search is performed. At least one of the information obtained from the search results by the condition
When a search process must be performed a plurality of times because it includes a designation step of designating one as a constraint condition and a setting step of setting a second search condition using the constraint condition designated in the designation step In addition, the search conditions for the next search process can be efficiently set using the previous search result, and the convenience of information search can be improved.

Further, according to the search condition generating method using the search result of the present invention (claim 4), after performing the search processing based on the set first search condition, the first search is performed. At least one of the information obtained from the search results by the condition
And a setting step of setting a second search condition using the constraint condition and the search condition specified in the specifying step. Therefore, when search processing must be performed multiple times, the previous search results can be set efficiently by using new search conditions to set search conditions for the next search processing. Convenience can be improved.

Further, according to the computer readable recording medium of the present invention (claim 5), claim 3 or 4
Since a program for causing a computer to execute each step of the search condition generation method using the search result described in the above is recorded, by executing this program on the computer, it is necessary to perform a plurality of search processes. , It is possible to efficiently set search conditions for the next search process using the previous search results, and realize a search condition generation method using search results that improves the convenience of information search can do.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram of an information search device according to an embodiment.

FIG. 2 is a schematic block diagram illustrating processing of a client illustrated in FIG. 1;

FIG. 3 is a schematic block diagram illustrating processing of a search server illustrated in FIG. 1;

FIG. 4 is a flowchart showing a process of generating a transposed file in the information search device of the present embodiment.

FIG. 5 is a flowchart showing a normal search process in the information search device of the present embodiment.

FIG. 6 is a flowchart illustrating a search process using a search result in the information search device of the present embodiment.

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 client 101 document DB 102 transposed file 103 search server 104 network 200 query 201, 300 natural language processing module 202, 301 dictionary 203, 302 grammar dictionary 204 constraint condition 205 search word 206 search condition 207 constraint condition setting unit 208 designation unit 303 document Set 304 Database Build Component 305 Query Build Component 306 Query Document 307 Search Engine 308 Document List

Claims (5)

[Claims] 1. An information search apparatus for performing a search process based on a specified search condition, wherein: a first setting means for setting a first search condition; and a first setting means for setting the first search condition. Search means for performing a search process based on search conditions; designating means for designating at least one of information obtained from a search result based on the first search condition as a constraint condition; constraint condition designated by the designating means Setting means for setting a second search condition by using the information search device, wherein the search means performs a search process based on the second search condition set by the setting means. . 2. An information retrieval apparatus for performing a retrieval process based on a designated retrieval condition, comprising: a first setting means for setting a first retrieval condition; and a first setting means for setting the first retrieval condition. Search means for performing a search process based on a search condition; designating means for designating at least one of information obtained from a search result based on the first search condition as a constraint condition and additionally designating a new search condition And setting means for setting a second search condition using the constraint condition and the search condition specified by the specifying means, wherein the search means sets the second search condition by the setting means. An information search device for performing a search process based on the information search device. 3. A specifying step of performing a search process based on the set first search condition, and then specifying at least one of information obtained from a search result based on the first search condition as a constraint condition. A setting step of setting a second search condition using the constraint condition specified in the specifying step, and a search condition generation method using a search result. 4. After performing a search process based on the set first search condition, at least one of information obtained from a search result based on the first search condition is specified as a constraint condition, and Using a search result characterized by including: a specifying step of additionally specifying a new search condition; and a setting step of setting a second search condition using the constraint condition and the search condition specified in the specifying step. How to generate search conditions. 5. A computer-readable recording medium recording a program for causing a computer to execute each step of the search condition generating method using the search result according to claim 3 or 4.