JPH0934911A - Information retrieval device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily presume the contents of a retrieved document and to quickly select a retrieval document by displaying the number of times of the appearance of the retrieval word of retrieved information for the respective documents. SOLUTION: A retrieval condition storage part 3 stores the retrieval conditions such as the retrieval word, derivation conditions, a sorting order and filtering conditions, etc., inputted from an input processing part 2 and a retrieval processing part 4 retrieves processing of the plural documents stored in a document storage part 1 corresponding to the retrieval word of the retrieval conditions stored in the storage part 3, the compound word of the retrieval word in a range derived by the derivation conditions and the relating word of the retrieval word. In this case, a retrieval word counting part 5 respectively counts the number of times of the appearance of the retrieval word appearing in the document by the retrieval processing, the compound word of the retrieval word in the range derived by the derivation conditions and the relating word of the retrieval word. A retrieved result storage part 6 stores the number of times of the appearance of the retrieval word, the compound word in the range derived by the derivation conditions and the relating word counted in the retrieval word counting part 5 along with the document name of the retrieved document as retrieved results.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information retrieval device,
In particular, when searching for a desired document from a document database or the like, the number of occurrences of each search word of the searched information is displayed for each document, and the content of the searched document can be easily guessed, The present invention relates to an information search device capable of quickly selecting a document.

[0002]

2. Description of the Related Art Conventionally, as a technique related to this type of information retrieval device, for example, "Document processing device with list output function" described in Japanese Patent Laid-Open No. 5-266022 and Japanese Patent Laid-Open No. 5-314181. "Information retrieval device" described in
It has been known.

The former "document processing device with a list output function" described in Japanese Patent Laid-Open No. 5-266022 discloses:
This is a proposal of a technique related to a technique for searching a plurality of documents, determining whether or not a search word is included, and outputting a registered name of a document including the search word as a registered name list.

The latter "information retrieval device" described in Japanese Patent Laid-Open No. 5-314181 is a proposal regarding a technique for retrieving the number of times a keyword input by a user is used in a target text. .

[0005]

By the way, when the technique disclosed in Japanese Patent Laid-Open No. 5-266022 is used for information retrieval, if the number of documents to be retrieved is small, the registered name Even if only the list is output, the load on the user does not become a problem, but if the number of documents to be searched increases, the load on the user increases.

[0006] The latter "information retrieval device" described in Japanese Patent Laid-Open No. 5-314181 is a method for narrowing down a search term in advance when a search is performed, and the number of search terms is complex for the user. The feature is that each word or synonym is presented, and words can be selected by a group before being searched. However, the number of cases corresponds only to words, and when multiple documents are searched, what kind of search word is included in what document, in what combination? Is not presented.

An object of the present invention is to display the number of occurrences of a search word of retrieved information for each document when a desired document is retrieved from a document database or the like, so that the content of the retrieved document can be easily guessed. Thus, it is an object of the present invention to provide an information search device capable of quickly selecting a search document.

[0008]

In order to achieve the above-mentioned object, the information retrieval apparatus which is the first feature of the present invention is
A document storage means (1) for storing a plurality of documents, an input means (2) for inputting a search word, a search word storage means (3) for storing an input search word, and a search word based on the input search word Search means (4) for searching for documents in the document storage means;
It is characterized by comprising counting means (5) for counting the number of appearances of the search word for each document and output means (6, 7) for outputting the counting result for each document.

In addition, in the information retrieval apparatus of the second aspect of the present invention, in addition to the above-mentioned configuration, a derivative word generating means (8) for generating a derivative word of the input search term.
The search means (4) performs a search including the derived words generated in the search word, and the counting means (5) includes the number of occurrences of each of the search word and the derived words in the document. Is counted for each document.

[0010]

In the information retrieval apparatus having the above characteristics, the document storage means stores a plurality of documents, and when a search word is input by the input means, the input search word is
It is stored in the search word storage means. The search means searches for a plurality of documents in the document storage means based on the input search word. At the time of the search, the counting unit counts the number of appearances of the search word for each document. Then, the output unit outputs the counting result for each document.

As a result, when the user selects a desired document from the retrieved results, the number of appearances of the retrieved word for each document is output as the counting result even if it is difficult to make a judgment only by the document name including the retrieved word. Therefore, the content of the document can be easily determined from the search word and the number of appearances of each document. For this reason, it becomes possible to select the document desired by the user at an earlier stage.

Further, in the information retrieval device which is the second feature of the present invention, further, a derivative word generating means is provided.
Derivative word generation means generates a derivative word from the input search word. The search means performs a search including the generated derivative word together with the search word, and the counting means counts the number of occurrences of each of the search word and the derivative word in the document for each document.

As a result, the number of appearances of each word in the document is output as the search result for each word of the search word and the derivative word for each document. Even if it is difficult to make a judgment based on the document name alone, the content of the document can be easily judged by referring to the number of appearances of derivative words, and the document desired by the user can be selected at an earlier stage. Become.

[0014]

Embodiments of the present invention will be specifically described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of an information search device according to an embodiment of the present invention. In FIG. 1, 1 is a document storage unit, 2 is an input processing unit, 3 is a search condition storage unit, 4 is a search processing unit, 5 is a search word counting unit, 6 is a search result storage unit, 7 is an output processing unit, 8 Is a derivative word generation unit, 9 is a derivative word storage unit, 10 is a compound word storage unit, 11 is a related word storage unit, 12 is a sorting processing unit, and 13 is a filtering processing unit.

The document storage unit 1 stores a plurality of document data to be searched. As will be described later (FIG. 2), each of the plurality of pieces of document data has a document name, a document attribute (attribute 1, attribute 2), and document content data.

As shown in FIGS. 7 to 9, which will be described later, the input processing unit 2 also displays an input screen in which an input operation by the user is displayed in a window, and search words, whether compound words are searched, and whether related words are searched. , Input processing that accepts search requests including search conditions such as derivation conditions, sort order, and filtering conditions. The search condition of the input search request is stored in the search condition storage unit 3.

As a search condition here, the input search word is a word to be searched, for example, a search word when performing a text search for a character string in document data,
It is a word contained in the document to be searched. The derivation condition is a condition that is derived from the search term in order to expand the range of the search process to the compound word and the related term. In this example, specifically, the derivation condition It is a condition for a compound word or a related word of a search word. Further, regarding the related word here, it is possible to arbitrarily select from a plurality of related words and input a setting (instruction of individual related word search) as a condition of the derived word. The sort order is a condition for defining the order in which the output processing unit 7 outputs the search result. The filtering condition is a condition for filtering the document data of the search output result when the output processing unit 7 outputs the search result.

The search condition storage unit 3 stores search conditions such as a search word, a derivation condition, a sort order, and a filtering condition input from the input processing unit 2. The search condition stored in the search condition storage unit 3, The search processing unit 4 performs a search process for a plurality of documents stored in the document storage unit 1 according to the compound words of the search words and the related words of the search words in the range derived by the derivation condition. In that case, the search word counting unit 5
By the search process, the number of appearances of the search word appearing in the document, the compound word of the search word in the range derived by the derivation condition, and the related word of the search word is counted.

The search result storage unit 6 includes, as the search result, the document name of the searched document, the search word counted by the search word counting unit 5, each compound word in the range derived by the derivation condition, and the related word. The number of appearances of is stored. That is, the number of appearances of the search word, the derivative word and the number of appearances of the derivative word, the attribute of the derivative word, and the like are stored for each document. Also,
An output flag is added to the search result. The output flag is a flag for instructing whether or not to output the document data of the search result data when the search result is filtered by the filtering condition. Output processing unit 7
Outputs the content of the search result storage unit 6 in accordance with the instruction to turn on or off the output flag.

Further, the derivative word generation processing unit 8 stores the compound word storage unit 10 and the related word storage unit according to the derivation condition of the search condition of the search condition storage unit 3 or the instruction from the input processing unit 2. The derivative word is generated by referring to the section 11, and the generated derivative word is stored in the derivative word storage unit 9. In this case, regarding the related words, only the related words selected by the input processing unit 2 are generated and stored.

The compound word storage unit 10 is a dictionary in which compound words are extracted and stored in advance from the document in the document storage unit 1, and the related word storage unit 11 stores related words such as similar words and synonyms. This is a dictionary. The sort processing unit 12 is a processing element that sorts the contents of the search result storage unit 6 based on the sort order of the search conditions of the search condition storage unit 3 when outputting the search results, and similarly, the filtering processing unit. Thirteen
When outputting the search result, the content of the search result storage unit 6 is filtered based on the filtering condition of the search condition of the search condition storage unit 3, and the output flag of the document data of the search result that is not output is set to “off”. Is a processing element.

Next, an example of the operation of the information retrieving apparatus constituted by each of the above elements will be concretely described. First, a specific example of a plurality of document data stored in the document storage unit 1 will be illustrated, and a case of performing a search for the document data will be described as an example.

FIG. 2 is a diagram showing an example of document data stored in the document storage unit. As shown in FIG. 2, each record of the plurality of document data 14 stored in the document storage unit 1 includes a document name field 14 a and an attribute 1 field 14
b, attribute 2 field 14c, and document content field 14d. The document name field 14a stores the document name of the document data, and the attribute 1 field 14b stores date data such as the creation date or the issue date of the document. Further, in the attribute 2 field 14c, source data such as the creator or issuer of the document is stored. In the document content field 14d,
Document content data is stored. In this way, as the document data, each data of the document name, the attribute, and the document content is stored. Attribute information such as the document creation date such as attribute 1 and the creator name such as attribute 2 is used here in the filtering process and the sorting process.

FIG. 3 is a diagram showing an example of compound word dictionary data stored in the compound word storage unit. As shown in FIG.
The compound word dictionary data stored in the compound word dictionary table 15 is data registered as a compound word by performing a morphological analysis on a document in the document storage unit 1 in advance and extracting a character string connected as a noun. is there. As such compound word dictionary data, compound word data from general kana-kanji conversion dictionary data may be used.

FIG. 4 is a diagram showing an example of derivative word data generated by the derivative word generation processing section and stored in the derivative word storage section. As shown in FIG. 4, a related word 17 and a compound word 18 are generated and stored as the derivative word data 16 stored in the derivative word storage unit. In the example of each word generated as the related word 17 and the compound word 18 here, the word “search” is input as the search word by the input processing unit 2 and the word generated by the derivative word generation processing unit 8 is input. Is an example of. As words related to “search”, three words “search”, “find”, and “search word” are generated and stored as related words, and “image search”, “image search”, “search management”, "Search word survey", ... Is generated (read) and stored as a compound word.

FIG. 5 is a diagram showing an example of the related word dictionary data stored in the related word storage unit. As shown in FIG. 5, the related word dictionary table 19 stored in the related word storage unit (11: FIG. 1) includes a word field 19a and a related word identification information field 19b. The word dictionary data is data in which the related word identification information is stored in the related word identification information field 19b corresponding to each word in the word field 19a. A word group in which the related word identification information is the same data means that the respective words are related to each other. More specifically, for example, the first word of the related word dictionary table 19
The related word identification information of the second word “search” is “A100”, and the related word identification information “A100” is extracted as the related word of the word “search”. , “Search word”, “search”, “search”, “search”, “search”, “search”, “find”, “search”, and “search” are obtained. Similarly, the related word identification information of the word "orange" is "B120", and the words "summer mandarin orange" and "mandarin orange" are obtained as the related words.

The order of the words in the related word dictionary data of the related word dictionary table 19 is to add readings of the words and arrange them in the order of reading, or, as shown in FIG. 5, in the related word identification information field 19b. Arrange in order of related word identification information. The data structure of these related word dictionary data may be a data structure that makes it easy to extract the related words related by the related word identification information here, as in the case of configuring general dictionary data.

FIG. 6 is a flow chart showing the overall processing flow of the information search processing according to the embodiment of the present invention. FIG.
The entire processing flow of the information search processing here will be described with reference to the flowchart of FIG. When the process is started, first, in step 20, a search request input process is performed. In the input process of the search request, the search condition is set and the start of the document search process is instructed. Next, in step 21, the document to be searched is read, and in the next step 22, the document is searched by the search word and the search word is counted for each document. And the next step 23
At, it is determined whether the count value is "0".

If the count value is not "0" in the determination of step 23, the process proceeds to step 24, and the retrieved document name, search word, count value, etc. are stored. Then, the process proceeds to step 25 to continue the search process for the next document. If the count value is "0" in the determination in step 23, nothing is done and the search process for the next document is continued as it is, so the process proceeds to step 25.

Next, in step 25, it is judged whether or not there are documents to be searched. If the document to be searched remains in this determination, the process returns to step 21, the process from step 21 is repeated, and the search process for the next document is continued. If there is no document to be searched for in the determination in step 25, next, in order to process the search result data, it is determined in step 26 whether or not the sort condition is ON. If the sort condition is ON, the process proceeds to step 27, the search results are sorted, and the process proceeds to the next step 28. If the sorting condition is not on, the process directly proceeds to the next step 28.

In step 28, it is determined whether the filtering condition is on. If the filtering condition is ON, the process proceeds to step 29, the search result is filtered, and the process proceeds to the next step 30. If the filtering condition is not on, the process directly proceeds to the next step 30. Then, in step 30, output processing is performed to output the processing results so far. In the output process in this case, only the document whose output flag is on “on” as a result of the filtering process is output as the search result.

Information retrieval processing is performed by such a series of processing flows. Next, an operation example of a retrieval request input when a user conducts information retrieval will be described. As described above, the operation of inputting the search request is performed on the user interface screen of the search request input window under the control of the input processing unit 2.

FIG. 7 is a diagram showing an example of an initial screen of the search request input window, and FIG. 8 is a diagram showing an example of a search request input window setting screen when instructing a compound word search and a related word search. Is. Further, FIG. 9 is a flowchart showing the processing flow of the search request input processing using the search request input window.

The search request input process by the search request input window here will be described with reference to FIGS. 7 to 9. When the user gives an instruction to input the search request, FIG.
As shown in, the initial screen of the search request input window 70 is displayed. A command menu bar 71 is provided at the top of the search request input window 70. The command menu bar 71 has a “close” button 72 for instructing the end of the setting of the search request input and a “instruction for starting the search process”. A "start" button 73 and a "stop" button 74 for instructing to stop this search processing are provided. Further, in the window, a search word input field 75 for inputting a search word, a compound word search instruction field 76 for instructing ON / OFF of the compound word search, and a related word search for instructing ON / OFF of the related word search An instruction field 77, a sort condition instruction field 78 that indicates use (on) / not use (off) of the sorting condition, and a filtering condition instruction field 79 that indicates use (on) / not use (off) of the filtering condition are provided. It is provided.

When the user makes a search request, the content settings of the search condition are input in these fields 75 to 79. For example, when the user desires to obtain a document including the word “search” and searches for a plurality of document data in the document storage unit, the search request input window 70
In, enter the word "search" as the search word, and enter the search request settings that instruct to turn on the compound word search, turn on the related word search, use the sorting condition (on), and use the filtering condition (on). To do. In this case, when each setting is input, the search request input window 70 changes to a display input screen 80 as shown in FIG. 8, and the search condition setting input of each search request is accepted.

In this case, as shown in FIG. 8, in the display input screen 80 of the search request input window in which the search condition setting input has been accepted, the search word input field 75 is displayed.
The character string 81 of the word “search” of the search word is input as text, and the compound word search instruction field 76 selects the ON menu item. Also, a related word search instruction field 7
In 7, the on-menu item is selected. In this case, when an ON menu item is selected in the related word search instruction field 77, a related word is generated for the word “search” of the previously input search word, and the related word display field 82 is displayed.
, The related words of each are displayed.

In the sort condition instruction field 78, when an ON menu item is selected and instructed, a sort condition setting field 83 for further instructing the contents of the sort condition is displayed. At 83, the user specifies the contents of the sort condition. Similarly, the filtering condition instruction field 79
When the ON menu item is selected and instructed, a condition setting field 84 for instructing the content of the filtering condition is further displayed. Therefore, the user specifies the content of the filtering condition in the condition setting field 84.

When such a search request is set and input and the "start" button 73 of the command menu bar 71 is pressed (positioned by the mouse pointer and the mouse button is clicked), the search processing here is started. . Next, the search request input process performed by the above-described operation will be described with reference to the flowchart in FIG.

When the processing is started, first, step 31
At, it is determined whether or not there is an input. The input determination process of step 31 is repeated until there is an input, and when there is an input, the process proceeds to step 32. Then, the determination process from step 32 to step 41 is repeated, the determination process of the input content is performed, and the process of step 42 to step 51 corresponding to each input content is performed according to the input content.

In the determination processing of the input contents from step 32 to step 41, first, at step 32, it is determined whether the compound word search button on is pressed. When the compound word search button on is pressed, the process proceeds to step 42, and the compound word search button o is displayed on the display screen.
A compound word search button on indicating that n has been selected is selected and displayed. Then, the process returns to step 31 again. When the compound word search button on is not pressed, the process proceeds to the next step 33.

Next, in step 33, it is determined whether the compound word search button off has been pressed. When the compound word search button off is pressed, the process proceeds to step 43, and the compound word search button off indicating that the compound word search button off is selected is displayed on the display screen. Then, the process returns to step 31 again. Compound word search button of
If f is not pressed, the process proceeds to the next step 34.

Next, in step 34, it is determined whether or not the related word search button on has been pressed. If the related word search button on is pressed, the process proceeds to step 44, and the related word search button on indicating that the related word search button on is selected is displayed on the display screen. In this case, the related word to be searched is displayed in the related word display field. Then, the process returns to step 31 again. If the related word search button on is not pressed, the process proceeds to the next step 35.

Next, in step 35, it is determined whether or not the related word search button off has been pressed. If the related word search button off is pressed, the process proceeds to step 45, and the related word search button off indicating that the related word search button off has been selected is displayed on the display screen. In this case, the display of the related word in the related word display field is erased. Then, the process returns to step 31 again. Related word search button of
If f is not pressed, the process proceeds to the next step 36.

Next, in step 36, it is determined whether or not the sort condition button on has been pressed. When the sort condition button on is pressed, the process proceeds to step 46, and the sort condition button on indicating that the sort condition button on is selected is displayed on the display screen. In this case, the sort condition setting field is displayed. Then, the process returns to step 31 again. If the sort condition button on has not been pressed, the process proceeds to the next step 37.

Next, in step 37, it is determined whether or not the sort condition button off has been pressed. If the sort condition button off has been pressed, the process proceeds to step 47, and the sort condition button off indicating that the sort condition button off has been selected is displayed on the display screen. In this case, the display of the sort condition setting field is deleted. Then, the process returns to step 31 again. Sort condition button of
If f is not pressed, the process proceeds to the next step 38.

Next, in step 38, it is determined whether or not the filtering condition button on has been pressed. If the filtering condition button on is pressed, the process proceeds to step 48, and the filtering condition button on indicating that the filtering condition button on has been selected is displayed on the display screen. In this case, the text input area of the condition setting field is displayed. Then, the process returns to step 31 again. If the filtering condition button on has not been pressed, the process proceeds to the next step 39.

Next, in step 39, it is determined whether or not the filtering condition button off has been pressed. When the filtering condition button off is pressed, the process proceeds to step 47, and the filtering condition button off indicating that the filtering condition button off has been selected is displayed on the display screen. In this case, the display of the text input area of the condition setting field is deleted. And
It returns to step 31 again. Filtering condition button o
If ff is not pressed, the process proceeds to the next step 40.

Then, in the next step 40, it is determined whether or not the character code is input. When the character code is input, the process proceeds to step 50 and the character code is set at the cursor position. Then, the process returns to step 31 again. If it is not a character code input, the next step 4
Go to 1 and determine whether the "start" button has been pressed. If the "Start" button is not pressed again,
Returning to step 31, the processing from step 31 is repeated. If it is determined that the "start" button has been pressed, the process proceeds to step 51, the search request content here is stored, and the process proceeds to the compound word search process.

Thus, in the search request input process,
In steps 32 to 39, it is determined that the on button or the off button of each search condition is pressed, and the selected state of each search condition button is displayed according to the content of the determination. If it is determined in step 40 that the character code has been input, the character code is input at the cursor position, and in step 41, it can be determined that the "start" button has been pressed. The request content is stored, and the process proceeds to the search process.

FIG. 10 is a diagram showing an example of search result data in the search result storage unit. As described above, when a search condition is set as a search request input and an instruction to start the search process is given from the search request input window (FIG. 8), the document data of a plurality of documents stored in the document storage unit (see FIG. The search process is executed for 2), and search result data (FIG. 10) having a data structure of a predetermined format is obtained in the search result storage unit. For example, as shown in FIG. 10, the search result data 60 here includes a document name field 61, an attribute 1 field 62, a search word appearance frequency field 63, a compound word field 64, a compound word appearance frequency field 65, and a related word field. 66, related word appearance frequency field 67, and output flag field 68 are record data of document data composed of data of respective fields,
The data in the fields of these record data are processed by sort processing and filtering processing, and the record data of the retrieved document data is output as the output result.

The search result data 60 shown in FIG. 10 shows an example of the search result after the count of the search words by the search word counting unit is completed. The document name, the attribute information (attribute 1), and the search for each document data are shown. Word occurrences, compound words and their occurrences,
The related word, the number of appearances thereof, and output flag data are included as field data. The data of the number of times of search word appearance in the number of times of search word appearance field 63 stores the number of times of appearance of the word “search” set as the search word for each document. In the attribute information (attribute 1) data of the attribute 1 field 62, the attribute data of the document data used in the sort condition or the filtering condition is stored as it is. In this example, the attribute information of the attribute 1 used in the filtering condition is stored. All output flags are "on".

Explaining in a concrete example, in the example of the search processing here, when the search processing is performed on a plurality of document data (FIG. 2) stored in the document storage unit, the search word in the search request input, Compound word search, related word search, sort condition,
Set the search condition consisting of each item of the filtering condition. For example, as shown in FIG. 8, the search term is “search”.
, Compound word search is on, related word search is on,
The sort condition "the number of times a search word appears" is set to on, and the filtering conditions "document creation date" and "number of documents" are set to on.

When the related word search is turned on in this search request input setting, the related words (search, search, search, search, search word, search, search, search) related to the word "search" previously input as the search word are searched. , Search, find) are displayed for each, so check the content of each related word.
If you select the filtering condition menu item (set it to "on"), a text area for entering the creation date and the number of documents will be displayed. input.

In this way, when the search request input is set and the start of the search process is instructed, the set search conditions are stored and the search process is started in accordance with the stored search conditions. When the search process is started, a derivative word is generated according to the search condition and is stored in the derivative word storage unit. Therefore, the derivative word is also included as the search word in the search process. In the case of this example, for example, from the word “search” input as a search word, among related words generated as a derivative word, for example, by performing an input operation of setting a search condition for search request input, , “Find”, “search word”, and “search” are stored as a compound word and a related word for the word “search” of the search word.

Next, the documents stored in the document storage unit are read one by one, and the document content data of the documents are retrieved. For example, the first document name “search device” is read, and the number of appearances of each word is counted while searching for the search word or derivative word for each read document. As a result, each word such as “search” of the search word stored as the search condition, the related word stored as a derivative word, the compound word of the search word and the compound word of the related word is searched, The number of appearances is counted for each word.

As a result, for example, the document name "search device"
“Search device”, “search speed”, “search word”, “search word search”, and the like appear in the document data of, and the count value of the number of appearances is obtained for each word. In addition, in the document data of the document name "multimedia search device", "image search", "multimedia search",
"Image search" and the like have appeared, and the count value of the number of appearances is obtained for each word. When the count value is obtained, the document data is output as the search result data, so the output flag is set to on.

The result of counting the search words for the document data of the document name "search device" is the result data 6 shown in FIG.
0 is shown as the first record data, and the result of the search word count for the document data of the document name “search device” is also shown as the second record data of the binding result data 60. ing. In the case of this search example, among the document data of a plurality of documents stored in the document storage unit, the document data of the document name “document creation device” and the document data of the document name “business show holding” are
Since the count value of the number of appearances of the search term was 0,
It is not stored as record data of the knot result data 60.

Search result data 6 thus obtained
With respect to 0, sorting is performed in the order of the sort condition set in the search condition of the search request input. In this example, since the “order of occurrence of search words” in the sort condition is “on”, the search result data 60 is searched according to the data content of the search word appearance count field 63 of the document data of the search result data 60. Record data of each document data of is sorted.

FIG. 11 is a diagram showing an example of search result data after sort processing in the search result storage unit. As a result of sorting by the sort condition in the search conditions of the search request input, the search result data 60 becomes the search result data 110 rearranged in the order of "the number of times of search word appearance" as shown in FIG. In the search result data 110 after the sorting process, the first document data is the document data of the document name “document search device” with the search word appearance count “58”, and the second document data is It is the document data of the document name "multimedia search device" whose number of appearances of the search word is "32". In the initial search result data 60, the document data of the document name “search device” located first in the search result data has a search word appearance count of “21” and is ranked 20th. It is sorted.

Next, with respect to the search result data 110 after the sorting process thus obtained, the document data to be output is further selected according to the filtering condition set in the search condition of the search request input. Perform filter processing. For a document that is not selected by the filter process and is not output, the output flag of the document data of the search result data is set to “off”.

FIG. 12 is a diagram showing an example of the search result data after the filter processing of the search result storage section. As a result of the filtering process based on the filtering condition of the search condition of the search request input and the search result data 110 after the sorting process,
As shown in 2, search result data 12 after filtering
It becomes 0. In this example, in the search request input (FIG. 8), the "document creation date and time" of the filtering condition is "January 1, 1993 to December 31, 1993", and the filtering condition Since the “number of documents” is designated as “top 20”, the filtering process is performed according to these conditions.

As a result of this filtering process, the output flag of the document data that does not meet the filtering condition is "of
f ”. In the search result data 110 after the sorting process shown in FIG. 11, the document data of the document name“ book search system ”and the document data of the document name“ Kana-Kanji conversion device ”have a small number of occurrences of the search word. This is an example of document data in which the output flag is set to "off" because it is not included in the top 20 in the sorting process. Similarly, in the document data of the document name “image search device”, the data of the creation date and time of the attribute 1 field 62 is “19”.
The output flag is set to "April 21, 1994" and does not match the range "January 1, 1993 to December 31, 1993" specified in the "document creation date and time" of the filtering condition. This is an example of document data that has been turned “off”.

In this way, the search result data 60 is sorted into the search result data 110 after the sorting process, and further, the filtering result is obtained as the search result data 120 after the filtering process, and the final search result is obtained. It becomes data and is output as search result output data. In the output process in that case, only the record data of the document data in which the output flag of the search result data is "on" is displayed and output as the search result data.

FIG. 13 is a diagram showing an example of output of search result output data. As shown in FIG. 13, in the output search result output data 130, each record data of the searched document data is associated with the document name 131, and the searched word / derivative data 132 is searched. The number of appearances 133 is output corresponding to each, and in accordance with the number of appearances of the search word, the record data of the 20 document data in the top 20 is displayed. Here, the attribute information (attribute 1, attribute 2) of the document data is displayed together so that the user can refer to it when selecting the retrieved document data.

The user (searcher) estimates the content of the document data while observing the displayed search result output data 130 and performs the work of determining the document data desired by the user. For example, assuming that the content of the document data desired by the user is related to the “natural language search method”, according to the search result output data 130, the document name “
The document data of "Multimedia search device" and the document name of document name "Document search method and device" are displayed in the upper position because they have many occurrences of search words, but they are derivative words (compound words, related words). Judging from the content of the number of appearances, it is possible to deduce that the document related to "language" does not appear and the document is unlikely to be related to the "natural language search method" desired by the user. .

In the search result output data 130, the derivative data (compound word, related word) is included in the document data of the first document name “document search device” and the document data of the twentieth document name “search device”. Word) and the number of appearances thereof are respectively displayed as “search word: 26 times” and “search word: 5 times”. Judging from these contents, the document data of the first document name “document search device” is displayed. It can be inferred that the document data with the 20th document name “search device” is more likely to be the desired document. Therefore, from the document data of the first document name “document search device” and the document data of the twentieth document name “search device”, it is presumed that it is wise to open the contents.

As described above, according to the information retrieving apparatus of this embodiment, a retrieval word and a word related thereto (a compound word including the retrieval word, a related word of the retrieval word, a relation (Compound word) is searched, and the number of occurrences in the search target document data can be compared. Therefore, even if the user cannot judge only by the document name, open it for each document and display its contents It is possible to infer the content of the document from the search word / derivative word (compound word, related word) and the number of appearances thereof, even without performing it. Therefore, it is possible to shorten the time for finding a desired document.

Next, a modified example of the information search apparatus according to the embodiment of the present invention will be described. In the above-described embodiment, when setting the search condition for inputting the search request (FIG. 8), if the menu item for turning on the related word search is selected and instructed, the related word is generated with respect to the previously input search word. Then, each related word is displayed in the related word display field, a search is performed for each of the related words, and a process of counting the number of appearances is performed. In that case, the generated related word is If the number is too large, the search process may be performed up to the related words that the user does not want, which may slow down the search process as a whole.

On the other hand, in the search request input,
The related word search is instructed, and the related word search is modified so that the related word search can be turned on / off for each related word in which the related word is generated. FIG. 14 is a diagram showing an example of a setting screen of a search request input window when the related word search is modified so that ON / OFF of each related word can be set, and FIG. 15 is a search request input of the modified example. It is a flowchart which shows the changed part of the process flow of the search request input process by a window.

The search request input process by the search request input window of this modification will be described with reference to FIGS. 14 to 15. Similar to the case described above, for example, when the user desires to obtain a document including the word "search" and performs document search, the word "search" is input as a search word, compound word search is turned on, and a related word is searched. When a search request is set by a search condition instructing search on, sort condition use (on), and filtering condition use (on), a search request input window is displayed as shown in FIG.
The display screen 90 receives the setting of the search request according to the search condition.

In this case, on the display screen 90 of the search request input window in which the search condition setting input has been accepted, the word "search" of the search word is entered in the search word input field 91.
Is input, the compound word search instruction field 92 is input.
Selects the on menu item. In the related word search instruction field 93, an on menu item is selected. In this case, when the ON menu item is selected in the related word search instruction field 93, a related word is generated for the previously input search word, and the related word display field 9 is displayed.
In 4, the related words of each are displayed.

In the search request input window of the modification shown in FIG. 14, when the related word search is set to “on” in the search condition setting input, the related word (search for the word “search” input as the search word (search) , Search, search, search, search, search word, find, search, search, find) are displayed as related word buttons 95 for instructing individual related word search for each related word. By using the related word button 95 of the instruction, a condition for searching for each related word for each related word is set. The state of the initial value is a state in which all the related word buttons 95 of the individual related word search instructions are “on”. This state is the same as the case described above. The setting of the other search conditions is the same as in the above case.

Next, the search request input process performed by the above-described operation will be described. Here, the processing step for instructing an individual related word search for each related word is added to the processing flow described with reference to FIG. That is, as shown in FIG.
In the flowchart partially showing the modified part of the search request input processing, the processing steps of steps 151 to 162 are the same as those of step 34 of the flowchart shown in FIG.
And is inserted between step 35 and step 35.

The process flow in this case will be described with reference to the flow chart of FIG. 15. The process flow up to step 34 and the process flow after step 35 are the same as those described with reference to FIG. The contents of the processing flow block 150 for performing the setting process of the conditions for the search of the individual related words for the newly added individual related words will be described.

As described above, when the process step of the search request input process proceeds and the process proceeds to step 34, step 34
In, it is determined whether or not the related word search button on has been pressed. If the related word search button on is pressed, the process proceeds to step 44, and the related word search button on indicating that the related word search button on is selected is displayed on the display screen. In this case, the related word to be searched is displayed in the related word display field, but here, it is displayed as an individual related word search instruction button. Then, the flow of processing returns to step 31, and it is determined whether or not the next input has been input. If the related word search button on is not pressed, next, step 1 in the process flow block 150
Go to 51.

In step 151, it is determined whether or not the related word 1 button for instructing the first individual related word search is turned on. When the related word 1 button is instructed to be on, the process proceeds to step 152, and the related word 1 button indicating that the search for the related word 1 is instructed is displayed on the display screen (selected display: reverse display, for example). And
The process proceeds to step 35 of the original processing flow. Also, related words 1
If the button is not instructed to be on, the process proceeds to the next step 153 as it is.

In step 153, it is determined whether the related word 1 button has been turned off. When the related word 1 button is instructed to be off, the process proceeds to step 154, and the related word 1 button indicating that the search for the related word 1 is not instructed is displayed off (non-selection display: for example, non-display) on the display screen. . Then, the process proceeds to step 35 of the original processing flow. If the related word 1 button is not turned off, the process proceeds to the next step 155 as it is.

In step 155, it is determined whether or not the related word 2 button for instructing the second individual related word search is instructed to be on. If the related word 2 button is instructed to be on, step 155 is performed. Proceeding to 156, the related word 2 button indicating that the search for the related word 2 is instructed is displayed on the display screen. Then, step 35 of the original processing flow.
Proceed to. If the related word 2 button is not instructed to be turned on, the process directly proceeds to the next step 157.

At step 157, it is determined whether or not the related word 2 button is turned off. When the related word 2 button is instructed to be off, the process proceeds to step 158, and the related word 1 button indicating that the search for the related word 1 is not instructed is displayed on the display screen. Then, the process proceeds to step 35 of the original processing flow. If the related word 2 button is not turned off, the third word
Step of determining whether or not the related word 3 button for instructing the th related individual related word search has been turned on (not shown)
Proceed to. The on / off determination processing of the related word n button is performed in the same manner until the determination of the related word (N-1) button.

Then, the process proceeds to step 159 for determining the next related word N button, and in step 159, it is determined whether or not the related word N button for instructing the Nth individual related word search is instructed to be on. To do. When the related word N button is instructed to be on, the process proceeds to step 152, and the related word N button indicating that the search for the related word N is instructed is displayed on the display screen. Then, the process proceeds to step 35 of the original processing flow. If the related word N button is not instructed to be on, the process directly proceeds to the next step 161.

Then, in step 161, it is determined whether or not the related word N button has been turned off. If the related word N button is turned off, step 162
Then, on the display screen, the related word N button indicating that the search for the related word N is not instructed is displayed off. Then, the process proceeds to step 35 of the original processing flow. If the related word N button is not turned off, the process directly proceeds to step 35 of the original processing flow.

Next, returning to step 35 of the original processing flow, it is determined whether or not the related word search button off has been pressed. If the related word search button off is pressed, the process proceeds to step 45, and the related word search button off indicating that the related word search button off has been selected is displayed on the display screen. In this case, the display of the related word in the related word display field is erased.

In this way, when the related word search is set to "on", a related word button (95) for instructing an individual related word search for all the related words of the related word input as the search word : Selection for Figure 14) /
The non-selection setting is performed, and the condition for each related word search for each related word is set. Then, only the related word search for the selected related word button is performed. As a result, it is not necessary to perform an unnecessary related word search process, so that the document search process can be performed efficiently.

FIG. 16 is a flow chart showing a changed part in the case where a related word is designated in the processing flow of the search request input processing and the process proceeds to the generation of the compound word for the compound word search. Similar to the above description with reference to FIG. 15, in the flowchart of FIG. 16, only changed portions are shown. The second modification will be described with reference to FIG.

Also in this case, step 4 here
Since the processing flow up to 1 is the same as that of FIG. 9 described above, the description thereof will be omitted, and the processing flow block 14 for performing the processing of generating and storing the newly added compound word here.
The contents of 0 will be described.

As described above, when the process step of the search request input process proceeds and the process proceeds to step 41, step 41
At, it is determined whether the start button has been pressed. If the start button has not been pressed, the process returns to step 31 and it is determined whether or not the next input has been input. If it is determined that the start button has been pressed, the process proceeds to step 141 of the process flow block 140, and the process of generating and storing the compound word for compound word search is performed.

In step 141, it is determined whether or not there is a request for compound word search. When there is no request for the compound word search, the process directly returns to the original processing flow. If it is determined in step 141 that there is a compound word search request, the process proceeds to step 142, and it is further determined whether or not there is a related word. If there is a related word, step 143
Then, a compound word of the related word is generated from the compound word storage unit, stored in the derivative word storage unit, and the process proceeds to the next step 144. If there is no related word, step 144 is left as it is.
Proceed to. Then, in step 144, a compound word of the search word is generated from the compound word storage unit, stored in the derivative word storage unit, and the original processing flow is returned to.

As described above, by partially modifying the processing flow of the search request input processing, when a related word is specified in the processing flow of the search request input processing, based on the specified related word, It can be modified so as to generate a compound word for compound word search. As a result, particularly for a related word that can be presumed to have a deep relationship, the search process of the present embodiment can be performed including a compound word based on the related word. The number of appearances of the compound word in the document is obtained. The user can more easily and more reliably estimate the content of desired document data from the number of appearances of the search word, the related word, and the compound word obtained as the search result data in the document.

The features of the embodiments and modifications of the information retrieval apparatus of the present invention described above can be summarized as follows. (1) In the information search device of the present embodiment, a document storage unit that stores a plurality of documents to be searched, an input unit that inputs a search word, and an input storage unit that stores the input search word. A search word counting unit that searches the document storage unit for a document including the search word stored in the input storage unit and counts the number of appearances of the search word in the searched document; A search result storage unit that stores the document name, each document, and the number of appearances, and an output unit that outputs the content stored in the search result storage unit. (2) Further, the system further comprises a derivative word generation unit that generates a derivative word of the search term stored in the input storage unit, and a derivative word storage unit that stores the derivative word generated by the derivative word generation unit, The search word counting unit searches for a document containing either a search word or a derivative word, and the search result storage unit stores the number of appearances of the search word or the derivative word for each document. (3) Further, a compound word storage unit for storing a compound word is provided, and in this case, the derivative word generation unit stores the compound word including the search word stored in the input storage unit in the compound word storage unit. Generate by referencing. (4) Further, the related word storage unit for storing a related word is further provided, and in this case, the derivative word generation unit stores the related word including the search word stored in the input storage unit as the related word storage. It is generated by referring to the section. (5) In the case of including two, a compound word storage unit that stores a compound word and a related word storage unit that stores a related word, the derivative word generation unit includes the search word stored in the input storage unit. The related word is generated by referring to the related word storage unit, and further, the compound word including the search word and the related word is generated by referring to the compound word storage unit. (6) Further, when outputting the contents stored in the search result storage unit, the output unit performs a sorting process or a filtering process and outputs the document attribute and the number of appearances of the search word for each document. (7) Further, the output unit outputs the counting result only for the documents counted as the number of appearances exceeding the predetermined threshold as a result of the sorting process or the filtering process.

[0090]

As described above, according to the information retrieval apparatus of the present invention, even when the user selects a desired document from the retrieved results, even if it is difficult to make a decision only by the document name including the retrieval word. Since the number of appearances of the search word for each document is output as the counting result, the content of the document can be easily determined from the search word and the number of appearances of each document. It becomes possible to select the document desired by the user at an earlier stage. Further, as the search result, the number of appearances in the document is output for each word of the search word and the derivative word for each document. Therefore, it is difficult to judge only by the document name including the number of appearances of the search word. Even in such a case, the content of the document can be easily determined from the number of appearances of the search word and the derivative word, and the document desired by the user can be selected at an earlier stage.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an information search device according to an embodiment of the present invention,

FIG. 2 is a diagram showing an example of document data stored in a document storage unit,

FIG. 3 is a diagram showing an example of compound word dictionary data stored in a compound word storage unit;

FIG. 4 is a diagram showing an example of derivative word data generated by a derivative word generation processing unit and stored in a derivative word storage unit;

FIG. 5 is a diagram showing an example of related word dictionary data stored in a related word storage unit;

FIG. 6 is a flowchart showing an overall processing flow of information search processing according to an embodiment of the present invention,

FIG. 7 is a diagram showing an example of an initial screen of a search request input window,

FIG. 8 is a diagram showing an example of a setting screen of a search request input window for instructing compound word search and related word search;

FIG. 9 is a flowchart showing a processing flow of search request input processing by the search request input window,

FIG. 10 is a diagram showing an example of search result data in a search result storage unit,

FIG. 11 is a diagram showing an example of search result data after sort processing in the search result storage unit,

FIG. 12 is a diagram showing an example of search result data after filtering in a search result storage unit;

FIG. 13 is a diagram showing an example of output of search result output data,

FIG. 14 is a diagram showing an example of a setting screen of a search request input window of the first modified example,

FIG. 15 is a flowchart showing a modified part of the processing flow of the search request input processing by the search request input window of the first modified example;

FIG. 16 is a flowchart showing a modified part of the search request input process of the second modified example.

[Explanation of symbols]

1 ... Document storage unit, 2 ... Input processing unit, 3 ... Search condition storage unit, 4 ... Search processing unit, 5 ... Search word counting unit, 6 ... Search result storage unit, 7 ... Output processing unit, 8 ... Derivative word generation Part, 9 ... derivative word storage part, 10 ... compound word storage part, 11 related word storage part, 1
2 sort processing unit, 13 ... filtering processing unit, 14 ...
Document data, 14a ... Document name field, 14b ... Attribute 1 field, 14c ... Attribute 2 field, 14d ... Document content field, 15 ... Compound word dictionary table, 16 ...
Derived word data, 17 ... Related word, 18 ... Compound word, 19 ... Related word dictionary table, 19a ... Word field, 19b ...
Related term identification information field, 60 ... Search result data, 6
1 ... Document name field, 62 ... Attribute 1 field, 63
... Search word appearance frequency field, 64 ... Compound word field, 65 ... Compound word appearance frequency field, 66 ... Related word field, 67 ... Related word appearance frequency field, 68 ... Output flag field, 70 ... Search request input window,
71 ... Command menu bar, 72 ... "Close" button, 73 ... "Start" button, 74 ... "Cancel" button, 7
5 ... Search word input field, 76 ... Compound word search instruction field, 77 ... Related word search instruction field, 78 ... Sort condition instruction field, 79 ... Filtering condition instruction field, 80 ... Display input screen, 81 ... Character string, 82
... Related word display field, 83 ... Sort condition setting field, 84 ... (Filtering) condition setting field,
90 ... Display screen, 91 ... Search word input field, 92 ...
Compound word search instruction field, 93 ... Related word search instruction field, 94 ... Related word display field, 110 ... Search result data, 120 ... Search result data, 130 ... Search result output data, 131 ... Document name, 132 ... Search word / Derived word data, 133 ... Number of appearances.

 ─────────────────────────────────────────────────── ─── Continued Front Page (72) Akio Yamashita Akio Yamashita 3-2-1 Sakado, Takatsu-ku, Kawasaki City, Kanagawa KSP R & D Business Park Building Fuji Xerox Co., Ltd. (72) Kazuo Aihara Sakado, Takatsu-ku, Kawasaki City, Kanagawa Prefecture 3-2-1 KSP R & D Business Park Building in Fuji Xerox Co., Ltd. (72) Inventor Tatsuomi Kita 3-2-1 Sakado, Takatsu-ku, Kawasaki City, Kanagawa Prefecture KSP R & D Business Park Building in Fuji Xerox Co., Ltd. (72) Inventor Yuko Matsuo 32-1 Sakado, Takatsu-ku, Kawasaki-shi, Kanagawa KSP R & D Business Park Building, Fuji Xerox Co., Ltd. (72) Inventor Shinji Kawamoto 3-2-1 Sakado, Takatsu-ku, Kawasaki-shi, Kanagawa KSP R & D Business Par Kubiru Fuji Xerox Co., Ltd. (72) Inventor Hiroshi Yamaguchi 3-2-1 Sakado, Takatsu-ku, Kawasaki City, Kanagawa Prefecture KSP R & D Business Park Building Fuji Xerox Co., Ltd.

Claims (2)

[Claims] 1. A document storage means for storing a plurality of documents, an input means for inputting a search word, a search word storage means for storing an input search word, and the document storage means based on the input search word. An information retrieval device comprising: a retrieval unit for retrieving the document, a counting unit for counting the number of appearances of the retrieval word for each document, and an output unit for outputting the counting result for each document. 2. The information search device according to claim 1, further comprising derivative word generation means for generating a derivative word of the input search word, wherein the search means is a derivative word generated as a search word. The information retrieving apparatus, wherein the counting unit counts the number of appearances of each of the retrieval word and the derivative word in the document for each document.