JP2001273329A - Method and system for retrieving information and recording medium with information retrieval processing program recorded - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately and efficiently retrieve information desired by a user. SOLUTION: A certain address is inputted as a retrieval request from among an address list obtained by browsing information in the past by a user himself or herself (a step S1). Whether or not the retrieval request is an address is decided (a step S2), and when it is an address, a document is obtained from a network based on the address (a step S3). Then, index generation processing is operated to the obtained document (a step S4), and the similarity of the index and the index generated for plurality of information which is already collected by a retrieval server is calculated (a step S5), and several documents whose similarity is high among the calculated similarity are outputted as a retrieved result (a step S6). Also, when the retrieval request is not an address but a text, a keyword, or file contents, the index is generated from them, and the processings in the step s5 and thereafter are operated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information retrieval method and an information retrieval system which enable efficient retrieval of information existing on a network, and a recording medium on which an information retrieval processing program is recorded.

[0002]

2. Description of the Related Art Search services for searching for information required by a user from a vast amount of information existing on a network are roughly classified into two types. One of them is
When a user inputs a keyword, there is a search service (herein called a keyword-type search) that searches for a set of documents that match the keyword and presents the searched set of documents to the user. A search service that reaches a desired document set by the user instructing and selecting a menu that has been hierarchically classified in advance and presents the document set to the user (this is called a directory-type search). There is.

Recently, there has emerged a service in which a concept of a certain information is compared with a concept of each information of an information group collected on a search server side, and a document having a similar concept of the input information can be searched. This search service is based on the above-described keyword search type as the search request content input by the user, and is capable of inputting a relatively long sentence (natural sentence) in the keyword portion, or a document searched by the search server. Some of those that are close to the user's request can be fed back to the search server. Here, what can be fed back to the search server is referred to as a feedback type search.

In this feedback-type search, in response to a search request given by a user in order to obtain some information, a search server presents some information to the user as a search result, and retrieves some of the searched documents. Is seen by the user,
The user selects a certain document from among them and tries to obtain a similar document, and feeds back the document to the search server. On the search service side, information that has already been collected for the document fed back from the user The document determined to be conceptually close to the document fed back from among the documents is presented to the user.

[0005]

In the above-mentioned feedback-type search, if a document set that is relatively close to a user's desired document is found in a set of searched documents, it can be fed back to a search server. Since the search process is performed again based on the fed-back document, it can be said that this is an effective method for obtaining information that the user really wants.

However, in many cases, there is no document to which feedback is desired in the search results presented from the search server. In some cases, the user already has an address (U) of information having a concept similar to the desired information.
RL), and may be trying to find information similar to that. In such a case, a document to be fed back is not always included in the information searched from the search server.

[0007] For example, when searching for information on a certain tourist spot, even if the search server searches for that tourist spot, the first search often fails to provide satisfactory information for the user. In such a case, a certain information is fed back from the search result to request the search again. At this time, if the user himself / herself knows some information about the sightseeing spot and wants to obtain detailed information related to the sightseeing spot, it is sufficient that the user himself / herself can feed back the information that he / she knows. Such information often does not exist in the first search result. In such a case, even if some information is selected from the search results and fed back, it is considered that after all, it is unlikely that the user can obtain the information that the user really wants.

As described above, in the conventional search system, it is difficult to efficiently obtain appropriate information even in the case of a feedback search.

SUMMARY OF THE INVENTION It is an object of the present invention to enable a user to accurately and efficiently search for information desired by a user when searching for information desired by the user from information collected by a search server.

[0010]

In order to achieve the above-mentioned object, an information retrieval method according to the present invention provides a method of retrieving predetermined information from information collected by a retrieval server based on a retrieval request given by a user. In an information search method for searching and presenting to a user, information having a concept close to the information desired by the user is input as a search request, and information based on the search request is selected from information collected by the search server. Is searched and output.

[0011] The information having a concept similar to the information desired by the user input as the search request includes the user's desired information selected from each of information in a group of information that the user has viewed in the past. Address of information having a concept similar to the information to be obtained, by giving the address as a search request, obtaining information based on the address from the network, and obtaining the obtained information and the respective information collected by the search server side. And the information to be presented to the user is determined from the similarity.

The information having a concept similar to the information desired by the user input as the search request includes:
At least one of sentences, keywords, and the contents of a file
By giving it as a search request, the content given as this search request is compared with each information collected on the search server side, and information to be presented to the user based on the similarity. It is also possible to determine

An information search system according to the present invention provides an information search system for searching for predetermined information from information collected by a search server based on a search request given by a user and presenting the information to the user. Search request input means capable of inputting information having a concept similar to desired information as a search request, and information from which a user selects and outputs information based on a search request from information collected by a search server And a search means.

In such a search system, the information input as the search request and having a concept close to the information desired by the user is defined as information from a group of information that the user has viewed in the past. An address of information having a concept close to the information desired by the selected user; in that case, an information acquisition means for acquiring information from a network is provided, and the address is given as a search request, thereby obtaining the information. Means for obtaining information based on the address from the network, comparing the obtained information with each information collected on the search server side,
Information to be presented to the user is determined from the similarity.

The information having a concept similar to the information desired by the user input as the search request includes:
At least one of sentences, keywords, and the contents of a file
By giving it as a search request, the content given as this search request is compared with each information collected on the search server side, and information to be presented to the user based on the similarity. It is also possible to determine

Further, the recording medium storing the information retrieval processing program of the present invention retrieves predetermined information from information collected by a retrieval server based on a retrieval request given by the user and presents it to the user. In a recording medium on which an information search processing program is recorded, the processing program inputs information having a concept close to the information desired by the user as a search request, and among the information collected by the search server, It includes a procedure for searching for and outputting information based on the search request.

In the recording medium on which such an information retrieval processing program is recorded, the information inputted as the retrieval request and having a concept similar to the information desired by the user includes:
An address of information having a concept similar to the information desired by the user selected from the information of each of the information groups that the user has viewed in the past, and by providing the address as a search request, Information based on the address is acquired from above, the acquired information is compared with each information collected on the search server side, and information to be presented to the user is determined based on the similarity.

The information having a concept similar to the information desired by the user input as the search request includes:
At least one of sentences, keywords, and the contents of a file
By giving it as a search request, the content given as this search request is compared with each information collected on the search server side, and information to be presented to the user based on the similarity. It is also possible to determine

As described above, according to the present invention, when a user attempts to retrieve some information, information having a concept similar to the information desired by the user is input as a retrieval request, and the information collected by the retrieval server is retrieved. From among them, information having a concept similar to the information as the search request given by the user is selected and presented to the user. Thus, information desired by the user can be obtained accurately and efficiently.

Then, as information having a concept similar to the information desired by the user input as the search request,
The user inputs an address for obtaining information having a concept close to the information desired by the user among the information of the information group viewed in the past. This is because a user who frequently browses various types of information from the Internet or the like in the past has an address (URL) for various types of browsed information in a personal computer (PC) as an address list. Often. Therefore, when searching for information having a concept similar to information that the user has viewed in the past, an address of information having a concept similar to the information desired by the user is given as a search request, and the address is given from the network. Information based on the address is obtained, the obtained information is compared with each information collected on the search server side, and information to be presented to the user is determined from the similarity. As a result, information desired by the user can be obtained accurately and efficiently.

The information having a concept similar to the information desired by the user input as a search request may be a natural sentence, a keyword input by the user, or a content existing in a certain file. This means that when a user tries to search for some information, the user inputs natural text, similar to the information to be searched, or as a keyword, and also inputs a certain file content as a search request. Thus, since information having a concept similar to the information given by the user is retrieved from the search engine, the information desired by the user can be obtained accurately.

[0022]

Embodiments of the present invention will be described below. The contents described in this embodiment are as follows.
The description includes the information search method and the information search system of the present invention, and also includes the specific processing contents of the information search processing program on a recording medium on which the information search processing program of the present invention is recorded.

According to the present invention, information having a concept close to the information desired by a user is input as a search request, and among the information already collected by the search server, the search request input by the user is selected. Information having a similar concept is selected and presented to the user.

At this time, the information input as the search request (information having a concept close to the information desired by the user) includes information desired by the user among information groups that the user has viewed in the past. The address of the information having a concept similar to the above is used.

In general, a user accustomed to information retrieval often frequently browses various kinds of information from the Internet in the past, and inputs addresses (URLs) for various kinds of browsed information in his / her PC. Often have a list of addresses. By designating an address in an address group present in the address list, information on the address present on the network can be obtained.

Further, the information search performed by the user often attempts to acquire information related to information acquired in the past. In such a case, from the address group existing in the address list obtained by browsing various information in the past, the address of the information likely to contain the content having a concept close to the information desired by the user is entered. By instructing, the information of the address present on the network is obtained, the obtained information is compared with the information already collected by the search engine, and some of the optimum information is selected from among them. It is presented to the user. As a result, information desired by the user can be obtained accurately and efficiently. Hereinafter, a specific description will be given.

FIG. 1 is a block diagram schematically showing the configuration of an information search system according to the present invention. In order to obtain information desired by a user (here, a document), a search request (contents of the search request) will be described. Is described later), an information acquisition unit 3 for acquiring a document from the network 2 as necessary based on the content of the search request input from the search request input unit 1, a search server 4 has an information search means 5 (selecting a document that is conceptually similar to the information input by the user as a search request) from the information group already collected and outputting the search result as a search result (its configuration will be described later). The configuration is as follows.

As described above, the search request input from the above-described search request input means 1 may be an address obtained by the user browsing various information in the past. For example, the sentence (natural sentence) input by the user or a keyword may be used.
Further, it may be the contents of a certain file. In addition,
When the search request given by the user is an address, the information acquisition unit 3 acquires a document from the network 2 based on the address. If the search request given by the user is a sentence (natural sentence) itself, a keyword, or even a file content,
Those contents are passed to the information retrieval means 5 through the information acquisition means 3.

The information search means 5 includes an index generation means 5
1. It has a similarity calculation means 52 and a search result output means 53. Further, the search server 4 has a collected information storage unit 41 that stores, as an index, each piece of information already collected from the network.

When an address is input as a search request from the user, the index generation means 51 generates an index for the document obtained by the information obtaining means 3 based on the address. When the text itself, a keyword, and the contents of a certain file are input, an index is generated for them. Then, the generated index and the above-mentioned collected information storage unit 4
1 is compared with the index for each piece of information, and the similarity calculating means 52 obtains the similarity. Based on the obtained similarity, for example, some documents are searched in descending order of similarity. Is output from the search result output means 53.

The operation procedure of the information retrieval system thus configured will be described with reference to the flowchart of FIG. First, it is assumed that the user has specified a certain address from a list of addresses obtained by browsing various information in the past from the search request input unit 1. That is, as described above, when an attempt is made to search for a document desired by the user, an address of information containing contents having a concept similar to that of the desired document is selected from an address list.

When a certain address is input as a search request (step s1), it is determined whether or not the user's search request is an address (step s2). To obtain a document from the network 2 (step s3). Thereafter, an index generation process is performed on the obtained document to generate an index for the document (step s4). Next, the similarity between the index generated for the acquired document and the storage content of the collected information storage unit 41 (index generated for many documents already collected by the search server 4) is determined. The degree is calculated by the degree calculating means 5 (step s5). Then, the documents are arranged in descending order of the similarity among the calculated similarities, and are output as search results (step s6). It should be noted that this search result can also include the similarity calculated for each document.

When acquiring a document from the network 2 based on the address given as the search request, a plurality of documents may be acquired. In this case, an index is generated for each of the acquired documents, and the contents stored in the collection information storage unit 41 (for a large number of documents already collected by the search server 4, Index). Furthermore, it is possible to provide not only one address but also a plurality of addresses as a search request. In this case as well, an index is generated for a plurality of types of documents obtained for each address, and the index and collection are performed. You may make it compare with the storage content of the information storage part 41.

In this way, the search result is presented to the user. In this example, an address of information that the user has viewed in the past is given as a search request, whereby the information is obtained from the network, and a document having a concept similar to the obtained information is searched by a search engine. Since such processing is performed, the search result presented to the user is likely to be a document desired by the user.

In the flowchart of FIG. 2, if the search request given by the user is not an address but a sentence itself, a keyword, or the contents of a certain file, the following processing is performed.

If any of the sentence itself, the keyword, and the content of a certain file is input as a search request, it is determined in step s2 of FIG. 2 that the address is not an address, and the input content is determined. (Step s4). For example, if a user inputs a sentence as a search request, an index is generated from the input sentence. Similarly, an index is generated from keywords and file contents input for keywords and file contents.

Then, as in the above, the processing after that is similar to that described above, the index generated for the input text, keyword, and file contents, and the storage contents of the collection information storage unit 41 (many data already collected by the search server 4). (The index generated for the document) is calculated by the similarity calculating means 5 (step s5), the documents are arranged in descending order of the calculated similarity, and are output as a search result. (Step s6).

According to this, the search results presented to the user are similar to the natural components and keywords given by the user and the contents of a certain file when the user tries to search for some information. Since the search server 4 retrieves the document that the user has, a document desired by the user can be obtained accurately.

It should be noted that a sentence or keyword given as a search request and / or a file content may be inputted in combination. Also in this case, an index is generated for each, and the similarity between the index and the storage content of the collection information storage unit 41 (index generated for a large number of documents already collected by the search server 4) is determined. Let's calculate.

The present invention is not limited to the embodiment described above, but can be variously modified without departing from the gist of the present invention. Further, the information search processing program for performing the processing of the present invention described above includes:
The information can be recorded on a recording medium such as a floppy disk, an optical disk, or a hard disk, and the present invention includes the recording medium. Further, the processing program may be obtained from a network.

[0041]

As described above, according to the present invention, when searching for information desired by a user, information having a concept close to the desired information is input as a search request and sent to the search server. Out of the information already collected
Since information having a concept close to the information given as the search request given by the user is selected and presented to the user, the information desired by the user can be obtained accurately and efficiently.

As the search request, an address of information having a concept close to information desired by the user can be used. This is effective when trying to search for information having a concept similar to the content of information that has already been browsed, obtains information based on the address from the network, and collects the obtained information and the search server. Then, the information to be presented to the user is determined based on the similarity. As a result, information desired by the user can be obtained accurately and efficiently.

Also, the search request includes a sentence,
You can also enter keywords or the content that exists in a file. This means that when a user tries to search for some information, the user inputs a content having a concept similar to the information to be searched in a natural sentence, as a keyword, and further, inputs a file content. Since a document having a similar concept to the information provided by the user is retrieved from the retrieval server, the information desired by the user can be obtained accurately and efficiently.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an information search system of the present invention.

FIG. 2 is a flowchart illustrating an information search processing procedure according to the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Search request input means 2 Network 3 Information acquisition means 4 Search server 5 Information search means 41 Collected information storage part 51 Index generation means 52 Similarity calculation means 53 Search result output means

Claims (9)

[Claims] 1. An information search method for searching for predetermined information from information collected by a search server based on a search request given by a user and presenting the predetermined information to a user. An information search method comprising: inputting information having a search request as a search request; and searching and outputting information based on the search request from information collected by the search server. 2. The information having a concept similar to the information desired by the user, which is input as the search request, is the information of the user selected from each information of the information group that the user has viewed in the past. An address of information having a concept similar to desired information, and by giving the address as a search request, information based on the address is obtained from the network, and the obtained information and the information collected by the search server are respectively obtained. 2. The information retrieval method according to claim 1, wherein the information to be presented to the user is determined based on the degree of similarity. 3. The information having a concept similar to the information desired by the user input as the search request is at least one of a sentence, a keyword, and the content of a certain file, and is given as the search request. 2. The information according to claim 1, wherein the content provided as the search request is compared with each information collected on the search server side, and information to be presented to the user is determined based on the similarity. retrieval method. 4. An information search system which searches for predetermined information from information collected by a search server based on a search request given by a user and presents the information to a user. A search request input unit capable of inputting information having a search request as a search request, and an information search unit searching and outputting information based on the search request from information collected by a search server. Characteristic information retrieval system. 5. The information having a concept close to the information desired by the user, which is input as the search request, includes the information of the user selected from each information of the information group that the user has viewed in the past. An address of information having a concept similar to desired information, in which case, an information acquisition unit for acquiring information from a network is provided, and the address is given as a search request, so that the information acquisition unit The information based on the address is obtained, the obtained information is compared with each information collected on the search server side, and information to be presented to the user is determined from the similarity. Information retrieval system. 6. The information having a concept similar to the information desired by the user input as the search request is at least one of a sentence, a keyword, and the content of a certain file, and is given as the search request. 5. The information according to claim 4, wherein the information provided as the search request is compared with each information collected on the search server side, and information to be presented to the user is determined from the similarity. Search system. 7. A recording medium storing an information search processing program for searching for predetermined information from information collected by a search server based on a search request given by a user and presenting the information to a user. Includes a step of inputting information having a concept close to the information desired by the user as a search request, and searching and outputting information based on the search request from information collected by the search server. A recording medium storing an information search processing program characterized by the following. 8. The information input as the search request and having a concept close to the information desired by the user is the information of the user selected from each information of the information group that the user has viewed in the past. An address of information having a concept similar to desired information, and by giving the address as a search request, information based on the address is obtained from the network, and the obtained information and the information collected by the search server are respectively obtained. 8. The recording medium according to claim 7, wherein the information to be presented to the user is determined based on the degree of similarity. 9. The information having a concept similar to the information desired by the user input as the search request includes text,
By giving at least one of a keyword and the contents of a certain file as a search request, the contents given as the search request are compared with respective information collected on the search server side, and the similarity is determined. The information to be presented to the user is determined from the following.
A recording medium on which the described information search processing program is recorded.