JP2008250893A - Information retrieval device, information retrieval method and its program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To retrieve optimal information for learning a teaching material from the question content of a learner on the teaching material. <P>SOLUTION: This information retrieval device 1 for retrieving the information for acquiring an answer to a question from a question sentence on predetermined first information acquired from a server; reads the question sentence (S11); extracts a first keyword by analyzing a morpheme (S12); reads a document of the first information; extracts a second keyword by analyzing the morpheme; executes weighting processing on the second keyword; determines a weight value of indicating significance of the respective second keywords; selects a keyword of high significance from the weight value: sets the selected keyword to an important keyword; generates a retrieval expression from a combination of the first keyword and the important keyword (S13); and outputs the document (a WEB page) for acquiring the answer to the question by the generated retrieval expression (S14). <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an information search apparatus, an information search method, and a program thereof, and in particular, in a WEB-based learning system that performs learning using an electronic teaching material, information that is optimal for learning the teaching material from the contents of a question of a learner That is, the present invention relates to an information search apparatus, an information search method, and a program for enabling search for information for obtaining an answer to a question.

  2. Description of the Related Art Today, with the expansion and advancement of computer networks, search engines that can search for desired information using keywords or the like from information published on the Internet when information is desired are widely used. Search engines store the full text of information published on the WWW or the like, and can be broadly classified into two types: a full text search type that searches by keywords and a directory type that is classified by category. Recently, a search function has been installed in a directory-type search engine.

  Even in a WEB-based learning system in which learning is performed using an electronic teaching material, a keyword search is essential for obtaining information. A WEB-based learning system generally has a function of accepting a question from a learner, but an answer to the question needs to be manually created by an instructor. For this reason, it costs and it takes time to obtain the answer. A search formula consisting of keywords that the learner can search and obtain answers by using a search engine on the WEB instead of the instructor creating an answer, but is input to the search engine to obtain desired information It is difficult to obtain an answer by properly creating

  In addition, in the WEB-based learning system, in order to keep the learner's motivation, an immediate response to the question point is necessary, and it is slow in the mail and the posted version. Immediateness can be maintained by using the chat, but in order to respond immediately to the chat user, the instructor must always be on standby, which is costly. A learner can search a file or a web page using a search engine to solve a question, but generally, a search engine user inputs information to the search engine to obtain desired information. It is difficult to create a search expression composed of keywords suitable for the search, and it is often the case that only one search keyword is input for search. In this case, a large number of WEB pages are searched at a time, and it is difficult to obtain a valid related WEB page.

  In addition, in the conventional WEB-based learning system, there is a technique for discriminating a category from a text sentence (document) input by a user and narrowing down a search result. However, preparation in advance is necessary and cost is increased.

  However, the information search system described in Patent Document 1 includes a search device that searches for information on an information network, a search request device that makes a search request to the search device based on the original text and a question from the user regarding the original text, In the information search system including the search device, the search device includes an original text specifying unit that specifies the original text, a keyword extracting unit that extracts the search keyword from the learner's question text, and a question that specifies the question attribute of the question based on the search keyword Specify the word attribute of the word that can be used to narrow down the range to be searched by the search device based on the attribute specification unit and the specified question attribute, extract the word belonging to the word attribute from the original sentence, and extract the extracted word And an additional key extractor that uses as an additional key to the search keyword, and when the additional key is extracted, the additional key is added to the search keyword. And, having an association unit to search device to create a search request based on the query, associating the search request to add key extracted from the original.

  In short, Patent Document 1 specifies a word attribute for narrowing a search range based on a question attribute specified based on a search keyword extracted from a question sentence, and extracts a word belonging to the word attribute from the original sentence. Then, a technique for performing a search by adding the extracted word to a search keyword is disclosed.

JP 2005-11149 A (see [Claim 1], paragraph numbers [0002], [0004], [0007] to [0009], [0014], [00015] of the specification and [FIG. 1] of the drawings) ).

However, since the information search system and the search request device disclosed in Patent Document 1 extract words belonging to word attributes from the original text and add the extracted words to the search keywords to obtain a search formula, Depending on the description of the sentence and / or the original sentence, an unimportant search keyword is required, and there is a problem that a desired keyword cannot be extracted and cannot be narrowed down sufficiently.
Further, the invention described in Patent Document 1 has a problem that an answer to a question cannot be obtained automatically without causing the learner to perform any operation or consciousness.

  The present invention has been made to solve the above-described problem, and provides an information search apparatus, an information search method, and a program therefor for searching for optimum information for answering the question from the contents of the question about the information, and more specifically. Specifically, it is an object of the present invention to provide an information search apparatus, an information search method, and a program for automatically searching for information sufficiently narrowed down optimal for learning of a learning material from the contents of a learner's question regarding the learning material.

  An information search program according to the present invention that achieves the above object generates information for searching for information related to a question sentence in accordance with a question sentence input from a computer on the learner side, and searches for the information using the keyword A search program, an output means for outputting a first learning area extracted from a learning material composed of a plurality of learning areas divided into a plurality of learning areas to the learner terminal, and receiving a question sentence from the learner A question sentence receiving means, a first keyword extracting means for extracting a first keyword for search by morphological analysis of the question sentence received by the question sentence receiving means, and the first learning region output from the output means Stored important keywords to be added to the first keyword for searching for information on the question sentence A second keyword extracting means for extracting the important keyword as a second keyword for searching with reference to a key word definition table; and searching for information on the question sentence using both the first keyword and the second keyword. It is made to function as a search means which performs.

  In the information search program, the computer device further performs a morphological analysis on each of a plurality of learning regions constituting the teaching material, and determines an importance from an appearance frequency and extracts a keyword extracted as the important keyword. It is made to function as an important word definition table production | generation means which produces | generates a word definition table.

  The information search method according to the present invention that achieves the above object generates information for searching for information related to the question sentence according to the question sentence input from the computer on the learner side, and searches for the information using the keyword. A search method, wherein the computer device outputs an output step of outputting a first learning area extracted from a learning material composed of a plurality of learning areas to a learner terminal, and accepts an input of a question sentence from the learner A question sentence receiving step, a first keyword extracting step for extracting a first keyword for search by morphological analysis of the question sentence received by the question sentence receiving means, and the first learning region output in the output step Of important keywords to be added to the first keyword for searching for information on the question sentence. A second keyword extracting step of referring to the important word definition table to extract the important keyword as a second keyword for search; and searching for information on the question sentence using both the first keyword and the second keyword. And a search step for performing.

  In the information search method, the computer device further performs a morphological analysis on each of a plurality of learning regions constituting the teaching material, and determines the importance based on the appearance frequency as the important keyword. An important word definition table generation step for generating a word definition table is performed.

  An information search apparatus according to the present invention that achieves the above object generates information for searching for information related to a question sentence in accordance with a question sentence input from a computer on a learner side, and searches for the information using the keyword. An output means for outputting a first learning area extracted from a learning material composed of a plurality of divided learning areas to a learner terminal, and a question sentence receiving means for receiving an input of a question sentence from the learner First keyword extraction means for extracting a first keyword for search by morphological analysis of the question sentence received by the question sentence reception means, and among the keywords in the first learning region output in the output step , Referring to an important word definition table storing important keywords to be added to the first keyword for searching for information on the question sentence. Second keyword extracting means for extracting a keyword required as a second keyword for search; and search means for searching for information on the question sentence using both the first keyword and the second keyword. It is characterized by.

  The information search apparatus further performs a morphological analysis on each of a plurality of learning regions constituting the teaching material, and generates the important word definition table using keywords extracted by determining importance from appearance frequencies as the important keywords. And an important word definition table generating means.

  According to the information search device, the information search method, and the program of the present invention, the learner can input a question sentence about the learning content of the learning material without being aware of the search formula, and in accordance with the learning content of the learning material. It is possible to automatically search for the optimum information that the learner really needs.

  FIG. 1 is a block diagram of an information search apparatus according to the present invention. The information search device 1 of the present invention shown above the one-dot chain line in FIG. 1 is a server. The server 18 as a search engine and a personal computer on the learner side (learner PC) shown below the one-dot chain line in FIG. ) 19 and a network 17 are communicably connected to each other and perform a WEB-based learning between the learner and the information retrieval apparatus 1. The information search apparatus 1 includes a CPU 11 connected to each other via a bus line 10, an input device 12 such as a mouse and a keyboard, which is a question sentence receiving means for receiving an input of a question sentence from a learner, a memory 13, a hard disk 14, The communication unit 16 communicates with the server 18 and the learner PC 19 via the display 15 and the network 17.

  The memory 13 stores a program for causing the CPU 11 to function as the first keyword extraction unit 131, the second keyword extraction unit 132, the search unit 133, and the keyword definition table generation unit 134.

  The communication unit 16 also has a function as an output unit that outputs a first learning area extracted from a learning material composed of a plurality of learning areas to a learner terminal (learner PC) 19.

The first keyword extracting means extracts a first keyword for search by performing morphological analysis on the question sentence received by the question sentence receiving means.
The second keyword extracting means refers to an important word definition table storing important keywords to be added to the first keyword for searching for information on the question sentence among the keywords in the first learning area output from the output means. An important keyword is extracted as a second keyword for search.

The search means searches for information related to the question sentence using both the first keyword and the second keyword.
The important word definition table generating means performs a morphological analysis on each of a plurality of learning areas constituting the teaching material, and generates an important word definition table using the extracted keywords by determining the importance from the appearance frequency.

  In the hard disk 14, the text file for teaching material 141 storing documents of learning materials to be learned, which the learner asks, and the document as independent words (nouns, pronouns, verbs, adjectives, adjective verbs, adverbs, conjunctions, and verbs). A morphological analysis dictionary file 142 used for dividing is stored.

  The information search apparatus 1 receives a plurality of documents including a main document to be learned from the server 18 via the communication unit 16 and the network 17 and stores them in the text file group 141 for teaching materials. The plurality of documents received from the server 18 may be directly written into the memory 13 after being received. In the embodiment of the present invention, a plurality of documents including a main document to be learned are stored in the server 18 and transmitted from the server 18 to the information search apparatus 1. However, in another embodiment, the plurality of documents are other than the server 18. Information transmitted from a computer or information created by itself may be stored in the information search apparatus 1.

  Next, the information retrieval apparatus 1 uses the morphological analysis dictionary file 142 to analyze each document, extract independent words, analyze each document, calculate the importance of the independent words, and calculate the importance of the independent words in each document. At least one independent word is selected from the degrees as an important keyword.

FIG. 2 is a flowchart showing a basic procedure for presenting an appropriate document for obtaining an answer to a question regarding a document to be learned.
In step S1, the question text relating to the learning target document shown in FIG. 3 input from the learner PC 19 via the communication unit 16 is read.

In step S2, the question sentence input in step S1 is subjected to morphological analysis and a keyword in the question sentence is taken out.
In step S3, a morphological analysis is performed on a document that the learner is currently referring to, and a keyword in the document is extracted. The currently referenced document is obtained by communicating with the learner PC 19 or read when reading the question sentence.

In step S4, a search expression composed of keywords input to the search engine to obtain desired information is obtained by combining the keywords extracted from the question sentence in step S2 and the keywords extracted from the document being referred to in step S3. Generate.
In step S5, a search is executed using the search formula generated in step S4, and document information (WEB page) for obtaining an answer to the question is output to the learner.

  FIG. 3 is a diagram showing a specific example of the display screen 30 in which the description content of the main document of the teaching material is displayed in the upper portion and the important keywords extracted are displayed in the lower portion, and FIG. 4 is the description content of the related document in the main document of the teaching material. It is a figure which shows the specific example of the display screen 40 which displays on the upper stage and displays the important keyword taken out in the lower stage. The documents shown in the upper part of FIG. 3 and FIG. 4 are related to teaching materials for introductory laws, and are received from the server 18 by the learner's operation of the learner PC 19 and the text file group 141 for teaching materials in the hard disk 14 of the information search device 1 Stored in The main document is related to intellectual property rights and the related document is related to antitrust law. Each document has a different description. In the following, in each document, the portion indicated by... Indicates that the description is omitted.

  3 and 4, the display screens 30 and 40 shown in the upper part are displayed on the left side of the lower part as keywords (nouns) obtained as a result of extracting nouns from independent words extracted by morphological analysis of each document and weights of the keywords ( (Importance level) is calculated, and on the lower right side, a high importance keyword selected from the calculated keyword importance levels is shown.

  FIG. 3 and FIG. 4 are teaching materials for introductory laws, respectively, showing the contents of the main document and related documents, and the morphological analysis results, that is, the keywords (nouns, weights) and important keyword extraction results. As for the weight, a weight value generally used in statistical processing such as tf / idf value is calculated and used for keywords (in this case, nouns) appearing in the document 31 shown in FIG. 3 and the document 41 shown in FIG. . A keyword having a high weight value is stored in the memory 13 as an important keyword.

  The weights shown in FIGS. 3 and 4 are obtained by calculating weights generally used in statistical processing such as tf / idf values for keywords (in this case, nouns) appearing in the main document and related documents. It is.

  FIG. 5 is a diagram showing a keyword table, (A) is a diagram showing a keyword table in the main document of the learning material shown in FIG. 3, and (B) is a keyword table in a related document of the main document of the learning material shown in FIG. FIG. In the example shown in FIGS. 5A and 5B, a keyword having a keyword weight of 6.0 or more is set as an important keyword. This important keyword is indicated by a value 1 in the figure, and an important keyword that is not regarded as an important keyword is indicated by a value 0. Therefore, this keyword table is referred to as an important word definition table.

  This important keyword is an important keyword that should be added to the first keyword in order to search for information related to the question sentence, among the keywords in the currently referenced document (first learning area) output from the communication unit 16 to the learner PC 19. is there. The second keyword extracting means extracts an important keyword as a second keyword for searching with reference to the important word definition table.

  The important word definition table generation means performs morphological analysis on each of the plurality of learning areas constituting the teaching material, and determines the importance from the appearance frequency so that a keyword having a high importance value can be extracted as an important keyword. Generate a word definition table.

  FIG. 6 is a flowchart showing a procedure for presenting a document in which important keywords suitable for obtaining an answer to a question regarding a document to be learned are added. Here, an appropriate document means a WEB page for obtaining an answer to a question regarding a predetermined document to be learned.

In step S11, the question text relating to the learning target document shown in FIG. 3 input from the learner PC 19 via the communication unit 16 is read.
In step S12, the question sentence input in step S11 is subjected to morphological analysis and a keyword in the question sentence is taken out.

In step S13, desired information is obtained by combining the keyword extracted in step S12 with the important keyword of the document to be learned and the document shown in FIG. In order to obtain a search expression, a search expression composed of keywords input to a search engine is generated. The currently referenced document is obtained by communicating with the learner PC 19 or read when reading the question sentence.
In step S14, a search is executed using the search formula generated in step S13, and a document (WEB page) for obtaining an answer to the question is output to the learner.

  FIG. 7 is a diagram illustrating a screen 70 where a learner inputs a question sentence while referring to a document related to a document to be learned and performs a morphological analysis of the question sentence. 7, the question text input by the learner while referring to the document shown in FIG. 4 in step S11 in FIG. 6 is shown in the upper part of FIG. 7, and the result of the morphological analysis of the question text in step S12 in FIG. The result of keyword extraction is shown on the lower right side of FIG.

  The learner is referring to the antitrust law document shown in FIG. 4 and wants to know examples of violations of the antitrust law. The keywords extracted from the question text are “violation”, “detection” and “ There are only three words "example". Even if these three words are used as search keywords, it is difficult to search for violations of antitrust laws or cases of detection.

  FIG. 8 is a diagram showing a search result searched using a keyword extracted from a question sentence. The search result screen 80 shown in FIG. 8 is a document information including a keyword extracted from a question sentence, “violation”, “detection”, and “example” of three words, including a case of violation of antitrust law or an example of detection. (WEB page) is insufficient.

  FIG. 9 is a diagram showing a screen 90 where the learner has acquired the optimum document for learning from the question content related to the document to be learned according to the present invention. In FIG. 9, the keywords extracted from the question text entered by the learner are shown in the upper left, and stored in advance in the memory 13 corresponding to the teaching material document opened when the question text was entered, that is, the document shown in FIG. An important keyword is shown in the upper right, and an example in which the keyword extracted from the question sentence and the important keyword are combined is shown in the lower part.

  As shown in FIG. 9, in step S13 of FIG. 6, a search expression is generated with AND conditions for “violation”, “detection”, “example”, “manufacturer”, “price”, “competition”, and “exclusive”. By doing so, it can be seen that documents including violations of antitrust laws or cases of detection can be obtained at the top of the search results.

  FIG. 10 is a diagram showing a search result obtained by searching using a keyword obtained by combining a keyword extracted from a question sentence input by a learner and an important keyword corresponding to a teaching material document opened when the question sentence is input. . The search result screen 100 shown in FIG. 10 is shown to be sufficient as document information (WEB page) including examples of violation of the antitrust law or detection.

  FIG. 11 is a flowchart showing a process of acquiring each important keyword in a plurality of documents using the information search apparatus shown in FIG. The processing in this flowchart is processing in a preliminary preparation stage that is performed before the learner inputs a question sentence and obtains an answer thereto. By performing this process in advance, the processing time after inputting the question text can be shortened.

  In step S101, a plurality of documents constituting the teaching material are registered. That is, a plurality of documents including a main document to be learned and related documents of the main document are received via the communication unit 16 by an input operation from the learner PC 19 and stored in the text file group 141 in the hard disk 14. The main document is preferably a document searched based on a keyword extracted from a question sentence. The related document of the main document is preferably a document searched based on a keyword extracted from the question sentence.

In step S102, each document stored in step S101 is subjected to morphological analysis and a keyword is extracted.
In step S310, the keyword table is referenced. The keyword table (important word definition table) described above with reference to FIG. 5 is a list of keywords (in this embodiment, noun keywords and weights of each keyword) and determination values as to whether or not each keyword is an important keyword. To display.

  When creating the keyword table, statistical processing such as tf / idf values is performed for each keyword appearing in each document, and the weight (importance) of the keyword is calculated. More specifically, the main document and a related document of the main document are compared, and an information gain or tf-idf value that is a scale indicating the weight for the appearing noun is calculated. The information gain and tf-idf value are larger for words (keywords) that appear only in that document, not in other documents. The tf-idf value is an index indicating the appearance frequency of a keyword when a keyword representing the document is extracted from the keyword extraction target document. In this embodiment, a commonly used tf-idf value is calculated and used as an index indicating the importance.

  In step S104, a keyword having a high importance is selected for each document and stored in the memory 13 as an important keyword.

  The embodiment of the present invention described above has been described by taking as an example the procedure of executing steps S1 to S5 in the flowchart shown in FIG. 2 and steps S11 to S14 in the flowchart shown in FIG. As an embodiment, for example, the processing of steps S101 to S104 in the flowchart shown in FIG. 11 is executed in advance, so that the processing of steps S101 to S104 is executed in advance. Since the morphological analysis and keyword extraction are executed in advance, the processing time of step S13 in the flowchart shown in FIG. 6 can be shortened.

(Appendix 1)
An information search program for generating a keyword for searching for information related to the question sentence according to a question sentence input from a learner's computer and searching for the information by the keyword,
Computer equipment,
Output means for outputting to the learner terminal the first learning area extracted from the learning material composed of a plurality of learning areas;
A question sentence receiving means for receiving input of a question sentence from a learner;
First keyword extraction means for extracting a first keyword for search by morphological analysis of the question sentence received by the question sentence reception means;
Of the keywords in the first learning area outputted from the output means, the important keywords are referenced with reference to an important word definition table storing important keywords to be added to the first keywords for searching for information on the question sentence. Second keyword extracting means for extracting as a second keyword for search;
Search means for searching for information relating to the question sentence using both the first keyword and the second keyword;
An information retrieval program characterized by functioning.

(Appendix 2)
Said computer device,
Important word definition table generating means for performing morphological analysis on each of a plurality of learning regions constituting the teaching material, generating the important word definition table using the keyword extracted by determining the importance from the appearance frequency as the important keyword,
The information search program according to appendix 1, wherein the information search program is made to function as:

(Appendix 3)
An information search method for generating a keyword for searching for information related to the question sentence according to a question sentence input from a learner's computer and searching for the information by the keyword,
Computer equipment
An output step of outputting to the learner terminal the first learning area extracted from the learning material composed of a plurality of learning areas;
A question acceptance step for accepting input of a question from a learner;
A first keyword extracting step of extracting a first keyword for search by performing morphological analysis on the question sentence received by the question sentence receiving means;
Of the keywords in the first learning area output in the output step, the important keywords are referenced with reference to an important word definition table storing important keywords to be added to the first keywords for retrieval of information on the question sentence. A second keyword extraction step of extracting as a second keyword for search;
A search step of searching for information related to the question sentence using both the first keyword and the second keyword;
An information retrieval method characterized by:

(Appendix 4)
The computer device is
An important word definition table generating step of generating a keyword definition table using the keyword extracted by performing morphological analysis for each of a plurality of learning regions constituting the teaching material, judging the importance from the appearance frequency as the important keyword,
The information search method according to supplementary note 3, wherein:

(Appendix 5)
An information search device that generates a keyword for searching for information related to the question sentence according to a question sentence input from a learner's computer and searches for the information using the keyword,
Computer equipment
Output means for outputting to the learner terminal the first learning area extracted from the learning material composed of a plurality of learning areas;
A question sentence receiving means for receiving input of a question sentence from a learner;
First keyword extraction means for extracting a first keyword for search by morphological analysis of the question sentence received by the question sentence reception means;
Of the keywords in the first learning area output in the output step, the important keywords are referenced with reference to an important word definition table storing important keywords to be added to the first keywords for searching for information on the question sentence. Second keyword extracting means for extracting as a second keyword for search;
Search means for searching for information related to the question sentence using both the first keyword and the second keyword;
An information retrieval apparatus comprising:

(Appendix 6)
The computer device is
Important word definition table generating means for performing morphological analysis on each of a plurality of learning regions constituting the teaching material, generating the important word definition table using the keyword extracted by determining the importance from the appearance frequency as the important keyword,
The information search device according to appendix 5, characterized by comprising:

(Appendix 7)
An information search device that is connected to a server that stores various information and searches for information for obtaining an answer to a question from a question sentence related to predetermined first information acquired from the server,
First extraction means for reading the question sentence received from the computer on the learner side and extracting a first keyword by performing morphological analysis;
Second extraction means for reading each document of the first information and the second information related to the first information, and extracting a second keyword by morphological analysis;
The second keyword appearing in each document is weighted, a weight value indicating the importance of each second keyword is obtained, a keyword having a high importance is selected from the weight value, and the selected keyword is assigned to each document. Setting means to set each of the important keywords for,
Search expression generation means for generating a search expression from a combination of important keywords for the document of the first keyword and the second information;
An information retrieval apparatus comprising:

(Appendix 8)
The document of the second information is a document searched based on the first keyword.
The information search device according to appendix 7.
(Appendix 9)
The document of the first information is a document searched based on the first keyword.
The information search device according to appendix 7.

(Appendix 10)
An information search method for searching for information for obtaining an answer to a question from a question sentence related to predetermined first information acquired from a server storing various information,
A first extraction step of reading the question sentence received from the computer on the learner side and extracting a first keyword by morphological analysis;
A second extraction step of reading each document of the first information and second information related to the first information, and extracting a second keyword by morphological analysis;
The second keyword appearing in each document is weighted, a weight value indicating the importance of each second keyword is obtained, a keyword having a high importance is selected from the weight value, and the selected keyword is assigned to each document. Setting steps for each important keyword for,
A search expression generation step of generating a search expression from a combination of important keywords for the document of the first keyword and the second information;
An information retrieval method characterized by comprising.

(Appendix 11)
An information search program for searching for information for obtaining an answer to a question from a question sentence regarding predetermined first information acquired from a server storing various information,
On the computer,
A first extraction step of reading the question sentence received from the computer on the learner side and extracting a first keyword by morphological analysis;
A second extraction step of reading each document of the first information and second information related to the first information, and extracting a second keyword by morphological analysis;
The second keyword appearing in each document is weighted, a weight value indicating the importance of each second keyword is obtained, a keyword having a high importance is selected from the weight value, and the selected keyword is assigned to each document. Setting steps for each important keyword for,
A search expression generation step of generating a search expression from a combination of important keywords for the document of the first keyword and the second information;
An information retrieval program characterized by causing

It is a block block diagram of the information search device of this invention. It is a flowchart which shows the basic procedure which presents a suitable document for obtaining the answer of the question regarding the document of learning object. It is a figure which shows the specific example of the display screen 30 which displays the description content of the main document of a teaching material on the upper stage, and displays the important keyword taken out on the lower stage. It is a figure which shows the specific example of the display screen 40 which displays the description content of the related document of the main document of learning materials on the upper stage, and displays the important keyword taken out on the lower stage. 4A and 4B are diagrams illustrating a keyword table, in which FIG. 3A is a diagram illustrating a keyword table in the main document of the teaching material illustrated in FIG. 3, and FIG. 4B is a diagram illustrating a keyword table in a related document of the main document of the teaching material illustrated in FIG. It is. It is a flowchart which shows the procedure of showing the document which considered the important keyword suitable for obtaining the answer of the question regarding the document of learning object. It is a figure which shows the screen of the place where the learner inputs a question sentence while referring to the document relevant to the document of learning object, and is performing the morphological analysis of a question sentence. It is a figure which shows the search result searched using the keyword taken out from the question sentence. It is a figure which shows the screen of the place where the learner is acquiring the optimal document for the learning from the question content regarding the document of learning object by this invention. It is a figure which shows the search result searched using the keyword taken out from the question sentence which the learner input, and the keyword which combined the important keyword corresponding to the document of the teaching material opened when the question sentence was input. It is a flowchart which shows the process which acquires each important keyword in a some document using the information search device shown in FIG.

Explanation of symbols

1 Information Retrieval Device 10 Bus Line 11 CPU
12 Input device 13 Memory 14 Hard disk 15 Display 16 Communication unit 17 Network 18 Server 30, 40, 70, 80, 90, 100 Display screen (screen)
131 First keyword extraction means 132 Second keyword extraction means 133 Search means 134 Important word definition table generation means 141 Text file group for teaching materials 142 Morphological analysis dictionary file

Claims (6)

An information search program for generating a keyword for searching for information related to the question sentence according to a question sentence input from a learner's computer and searching for the information by the keyword,
Computer equipment,
Output means for outputting to the learner terminal the first learning area extracted from the learning material composed of a plurality of learning areas;
A question sentence receiving means for receiving input of a question sentence from a learner;
First keyword extraction means for extracting a first keyword for search by morphological analysis of the question sentence received by the question sentence reception means;
Of the keywords in the first learning area outputted from the output means, the important keywords are referenced with reference to an important word definition table storing important keywords to be added to the first keywords for searching for information on the question sentence. Second keyword extracting means for extracting as a second keyword for search;
Search means for searching for information relating to the question sentence using both the first keyword and the second keyword;
An information retrieval program characterized by functioning. Said computer device,
Important word definition table generating means for performing morphological analysis on each of a plurality of learning regions constituting the teaching material, generating the important word definition table using the keyword extracted by determining the importance from the appearance frequency as the important keyword,
The information search program according to claim 1, wherein the information search program is made to function as: An information search method for generating a keyword for searching for information related to the question sentence according to a question sentence input from a learner's computer and searching for the information by the keyword,
Computer equipment
An output step of outputting to the learner terminal the first learning area extracted from the learning material composed of a plurality of learning areas;
A question acceptance step for accepting input of a question from a learner;
A first keyword extracting step of extracting a first keyword for search by performing morphological analysis on the question sentence received by the question sentence receiving means;
Of the keywords in the first learning area output in the output step, the important keywords are referenced with reference to an important word definition table storing important keywords to be added to the first keywords for retrieval of information on the question sentence. A second keyword extraction step of extracting as a second keyword for search;
A search step of searching for information related to the question sentence using both the first keyword and the second keyword;
An information retrieval method characterized by: The computer device is
An important word definition table generating step of generating a keyword definition table using the keyword extracted by performing morphological analysis for each of a plurality of learning regions constituting the teaching material, judging the importance from the appearance frequency as the important keyword,
The information retrieval method according to claim 3, wherein: An information search device that generates a keyword for searching for information related to the question sentence according to a question sentence input from a learner's computer and searches for the information using the keyword,
Output means for outputting to the learner terminal the first learning area extracted from the learning material composed of a plurality of learning areas;
A question sentence receiving means for receiving input of a question sentence from a learner;
First keyword extraction means for extracting a first keyword for search by morphological analysis of the question sentence received by the question sentence reception means;
Of the keywords in the first learning area output in the output step, the important keywords are referenced with reference to an important word definition table storing important keywords to be added to the first keywords for retrieval of information on the question sentence. Second keyword extracting means for extracting as a second keyword for search;
Search means for searching for information relating to the question sentence using both the first keyword and the second keyword;
An information retrieval apparatus comprising: Important word definition table generating means for performing morphological analysis on each of a plurality of learning regions constituting the teaching material, generating the important word definition table using the keyword extracted by determining the importance from the appearance frequency as the important keyword,
The information search apparatus according to claim 5, further comprising:

Images