JPH0991311A - Information storage and retrieval device and its control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the information storage and retrieval device which can be improved in the efficiency of document registration and document retrieval and its control method by analyzing a document, informing an operator of a word that seems to be a new word or coined word and efficiently registering it in a word dictionary, and generating an index. SOLUTION: When a word registered in the word dictionary is extracted from the inputted document and a coined word other than the extracted word is extracted for characters in the document, the extracted coined word is registered in a coined- word candidate list. When the operator instructs a new-word registering process, the new word registered in the new word candidate list is displayed in a new word candidate list and when the marks '○' and '×' in a 'choice' area R21 are specified by mouse clicking, etc., the selection state is inverted each time the mouse is clicked and whether or not the new word is registered is selected. The area R24 is an area where new words in KATAKANA (square form of Japanese syllabary) are all selected together and the selection state of the KATAKANA new words can be set to '○'more easily than the marks '○' and '×' of the respective words in the area R21 are clocked.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information storage and retrieval apparatus for extracting a word from document information by a word dictionary and constructing an index for retrieval, and a control method therefor.

[0002]

2. Description of the Related Art A conventional information storage / retrieval device extracts a word from document information in accordance with a word dictionary and creates an index which is secondary information of the document information to improve the document retrieval efficiency. .

[0003]

However, in the above-mentioned conventional information storage and retrieval apparatus, when an operator registers a document, a new word or coined word such as a product name included in the document must be newly added to the word dictionary. I had to read through the document and check for new words and coined words in order to do so.
Such a word is often used as a keyword when searching a document, and if the index is not created because the word is not registered in the word dictionary, the search capability of the system is significantly reduced.

The present invention has been made in view of the above problems, and analyzes a document, notifies an operator of a word that is considered to be a new word or a coined word, efficiently registers the word in a word dictionary, and creates an index. Accordingly, it is an object of the present invention to provide an information storage / retrieval device and a control method thereof that can improve the efficiency of document registration and document retrieval.

[0005]

In order to achieve the above object, an information storage / retrieval apparatus of the present invention comprises a word dictionary for storing known words and an extracting means for extracting the words stored in the word dictionary from document information. And an index accumulating unit that adds an index for searching the extracted word to the word, in the document information, a set of characters other than the extracted word is targeted. A new word determining means for determining whether or not a character belonging to the set is a candidate for a new word, and a new word candidate character storage means for storing a character as a candidate for a new word as a result of the determination.
A new word selecting means for selecting a new word from the stored new word candidate characters according to an instruction of an operator; and a word dictionary registering means for registering the selected new word in the word dictionary as a known word. Characterize.

Preferably, the new word selecting means includes katakana character batch selecting means for collectively selecting katakana characters from the stored new word candidate characters.

Further, the control method of the information storage and retrieval apparatus of the present invention includes a word dictionary for storing known words, an extracting means for extracting words stored in the word dictionary from document information, and the extracted words. In the control method of the information storage and retrieval apparatus, further comprising an index assigning unit that assigns an index for searching the word, in the document information, a set of characters other than the extracted word is targeted, and the set It is determined whether the character belonging to is a candidate for a new word, and as a result of the determination, the character that is a candidate for the new word is stored in the new word candidate character storage means, and the stored new word candidate character A new word is selected according to an instruction, and the selected new word is registered in the word dictionary as a known word.

[0008]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a block diagram showing a schematic configuration of an information storage / retrieval device according to an embodiment of the present invention.

In FIG. 1, reference numeral 1 is a CPU that controls the entire information storage and retrieval apparatus of this embodiment.
Reference numeral 1 is connected to the display 2, the command input device 3, the ROM 4, the RAM 5, the secondary storage device 6, and the document input device 7 via the bus 8 so that data can be transmitted and received mutually.

The display 2 is composed of, for example, a CRT or a liquid crystal, and displays the information and options being created.

The command input device 3 is composed of, for example, a keyboard and a mouse, and is used for instructing various processes, moving a point of interest, selecting operation, inputting characters and symbols, and the like.

The ROM 4 is a read-only memory for storing various processing programs executed by the CPU 1, and is a partial area of the document registration program section 4a.
Stores programs and data for registering documents and outputting new word candidates.

The RAM 5 is a readable / writable memory that temporarily stores various calculation results and input information, and a new word candidate list section 5a which is a part of the area is a new word candidate list output by the CPU 1. Is used as a temporary storage area.

The secondary storage device 6 is composed of, for example, a hard disk, an MO (Magneto Optical disk), or the like.
Document information areas 6a and 2 for storing document information that is the next information
It has an index area 6b for storing an index which is the next information and a word dictionary area 6c for storing a word dictionary.

The document input device 7 is composed of, for example, an OCR document reading device and a communication device, and is used for inputting a document to be stored in the information storage / retrieval device of the present embodiment.

FIG. 2 is a diagram showing an example of a document input from the document input device 7 and stored in the document information area 6a.

FIG. 3 is a diagram showing only a part of the words stored in the word dictionary area 6c. As shown in the figure, a known word is stored in the word dictionary area 6c, and further information on whether or not the word is a keyword to be used for a search is added. When the keyword word is extracted from the document, the storage location of the document and the position where the keyword appears are stored in the index area 6b.

FIG. 4 is a diagram showing an example of a new word candidate list to which information of new word candidates extracted from some documents is added.

The control process executed by the CPU 1 of the information storage / retrieval device configured as described above will be described below with reference to FIGS.

FIG. 5 is a flow chart showing the procedure of the control processing executed by the CPU 1, and the program shown in this flow chart is stored in the ROM 4a of FIG.

When the operator turns on the power of the system, this control processing program is called, and first the process proceeds to step S1 to perform the necessary initialization processing.

Next, in step S2, a command input from the keyboard 3 is waited, and when any command is input, the process proceeds to step S3.

In step S3, the command is evaluated and the process branches depending on the content of the command. At this time, if a document is newly registered in the secondary storage device 6,
Proceed to step S4.

In step S4, the registered document is input from the document input device 7, and in step S5, the document is collated with the word in the word dictionary area 6c to extract the word contained in the document. FIG. 8 is a diagram showing the states of the extracted words in an easy-to-understand manner. In the figure, the character strings enclosed by () indicate the extracted words. The extracted words output as a result are used to create an index in step S8 described later.

Next, in step S6, a new word extraction processing subroutine for performing new word extraction processing is executed.

FIG. 6 is a flow chart showing the detailed procedure of this new word extraction processing subroutine, and FIG. 9 is a diagram showing in an easy-to-understand manner how a new word is extracted by this processing. In FIG. 9, a slash "/" represents a word delimiter. The new word extraction process will be described in detail below using this flowchart.

In FIG. 6, first in step S61,
A pointer to the document is added to the new word candidate list 5a. The new word candidate list 5a is a FIFO (first in first out)
Since it has the property of a queue and information is added to the end of the list, information can be retrieved in the order of addition by obtaining information from the beginning of the list. This is because when a certain new word is extracted as a candidate, the index for that word has not been created yet, and it is one of the operations to be performed later to identify the document that needs to be indexed after it is decided as a new word. It is a realization method.

Next, in step S62, the remaining character strings from which words have been extracted are narrowed down to the target. It is a character string that may include a new word.

Next, in step S63, words are delimited by boundaries of different character types such as kanji, hiragana, katakana, numbers, alphabets, and symbols, and blank characters.

Next, in step S64, the hiragana, the word consisting of one or less characters, and the word consisting of numbers are removed from each of the separated words.

The new word candidates extracted in this way are
In step S65, the words are rearranged in the order of spelling, duplicate words are eliminated in step S66, and the words are added to the new word candidate list 5a in step S67.

Returning to FIG. 5, when the new word extracting process in step S6 is completed, the process proceeds to step S7, and the document information is stored in the primary information area 6a in the secondary storage device 6.

Thereafter, the process proceeds to step S8, and step S
An index in which the word extracted in step 5 and the registration location of the document are associated with each other is constructed, and the process ends.

When the process of step S8 is completed, the process returns to the command waiting state of step S2.

As a result of the command evaluation in step S3, if the command is a new word registration command, the process proceeds to step S9, and a new word registration processing subroutine for performing new word registration processing is executed.

FIG. 7 is a flowchart showing the detailed procedure of this new word registration processing subroutine. The new word registration processing will be described in detail below with reference to this flowchart.

In the figure, first, in step S91, the presence or absence of the contents of the new word candidate list 5a is evaluated, and if the new word candidate list 5a has just been cleared, step S92 is performed as usual.
Go to and enter the word to register.

FIG. 10 is a diagram showing an example of an input screen by the GUI (graphical user interface) of the window system displayed on the display 2 in step S92. The command input device 3 is used to specify and input a region of the window by clicking the mouse. In the figure, a region R11 is a region for inputting a word to be registered, and when the region R12 is designated, the region R1
The word entered in 1 is determined, and the process proceeds to the next step.

Returning to FIG. 7, the flow advances to step S93 to register in the word dictionary.

If it is determined in step S91 that the new word candidate list 5a has contents, the process proceeds to step S94.

In step S94, the new word candidate list 5a
The word to be registered is displayed and the operator is asked whether or not to register. At this time, the operator need only specify whether or not to register each word.

FIG. 11 is a diagram showing an example of an input screen by the GUI of the window system displayed on the display 2 in step S94. Input and execution are performed by designating a window area from the command input device 3 by a method such as mouse clicking. In FIG.
Whether or not to register a new word is indicated in the area R21 by a mark XX, and the selection state is reversed by designating this mark by mouse click or the like. Area R22
And R23 are scroll buttons, which are areas for instructing to move the display range when a new word does not fit in the display area. When the area R22 is pressed, the display range moves upward, and when the area R23 is pressed, the display range moves downward. The region R24 is a region for instructing to collectively select new words in katakana, and the selection state of new katakana words can be set to ◯ more easily than clicking the ◯ × mark of each word in the region R21. The area R25 is an area for instructing to select all the new words in the list at once, and the selection state of all the new words in the list is more easily selected than clicking the ○ × mark of each word in the area R21. Can be When the region R26 is specified, the word in the region R21 having a selection state of x is deleted from the new word candidate list 5a, and the process proceeds to the next step.

Returning to FIG. 7, when the selection work is completed, the process proceeds to step S95, and the selected word is registered in the word dictionary.

Next, in step S96, the index of the related document is created. The related document is specified based on the pointer information to the document added in step S3 to the new word candidate list 5a. It is checked whether or not the new word is included in the documents added after that including the document in which the new word is extracted. If the new word is included, it is added to the index.

Next, in step S97, the new word candidate list 5a for which the processing has been completed is cleared and the processing ends.

Returning to FIG. 5, when the process of step S9 ends, the process returns to the command waiting state of step S2.

As a result of the command evaluation in step S3,
If it is any other command, the process proceeds to step S10, and after the necessary processing of each command is performed, the process returns to the command waiting state of step S2.

As described above, in the present embodiment, the input document is analyzed, characters that are not in the word dictionary, that is, words that seem to be new words or coined words are registered in the new word candidate list 5a, and the registered new words are registered. Is displayed in the new word candidate list, and the operator registers the desired new word from the displayed new word candidates by a method such as mouse click to register it in the word dictionary and is configured to add an index to this new word. , The new word can be efficiently registered in the word dictionary, and the efficiency of the document search can be improved by creating the index.

Further, when registering a new word, the new words of katakana characters can be selected all at once, so that the new word can be registered more easily.

The present invention may be applied to a system constituted by a plurality of devices or to an apparatus constituted by a single device. Further, it goes without saying that the present invention can be applied to the case where it is achieved by supplying a system or a program. In this case, by reading a storage medium storing a program represented by software for achieving the present invention into the system or the apparatus, the system or the apparatus can enjoy the effects of the present invention.

[0052]

As described above, according to the present invention,
With respect to the set of characters other than the words extracted by the extraction means in the document information, it is determined whether or not the characters belonging to the set are candidates for a new word, and as a result of the determination, a candidate for a new word is determined. Is stored in the new word candidate character storage means,
A new word is selected from the stored new word candidate characters in accordance with an instruction from the operator, and the selected new word is registered in the word dictionary as a known word, so that the word dictionary can be efficiently registered. Has the effect of enabling. Also,
When a new word is registered in the word dictionary, an index is added to the new word, so that it is possible to improve the efficiency of document search.

Further, when selecting a new word, the katakana characters among the stored new word candidate characters are collectively selected, so that the new word can be registered more easily.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an information storage / retrieval device according to an embodiment of the present invention.

2 is a diagram showing an example of a document input from the document input device of FIG. 1 and stored in a document information area.

FIG. 3 is a diagram showing a small part of words stored in a word dictionary area of FIG. 1;

FIG. 4 is a diagram showing an example of a new word candidate list to which information of new word candidates extracted from some documents is added.

5 is a flowchart showing a procedure of control processing executed by the CPU of FIG.

FIG. 6 is a flowchart showing a detailed procedure of a new word extraction processing subroutine of step S6 of FIG.

FIG. 7 is a flowchart showing a detailed procedure of a new word registration processing subroutine of step S9 of FIG.

FIG. 8 is a diagram showing the states of words extracted in step S5 of FIG. 5 in an easy-to-understand manner.

FIG. 9 is a diagram showing in an easy-to-understand manner a new word is extracted in the new word extraction processing of FIG.

10 is a diagram showing an example of a GUI input screen displayed on the display of FIG. 1 in step S92 of FIG.

11 is a diagram showing an example of a GUI input screen displayed on the display of FIG. 1 in step S94 of FIG.

[Explanation of symbols]

1 CPU (extracting means, index assigning means, new word judging means, new word selecting means, word dictionary registering means, katakana character collective selecting means) 2 Display (new word selecting means, katakana character collective selecting means) 3 Command input device (new word selecting means) , Katakana character batch selection means) 5a New word candidate list (new word candidate character storage means) 6c Word dictionary

Claims (3)

[Claims] 1. A word dictionary for storing a known word, an extracting means for extracting a word stored in the word dictionary from document information, and an index for searching the word for the extracted word. In an information storage / retrieval device provided with index assigning means, in the document information, for a set of characters other than the extracted words, whether or not the characters belonging to the set are candidates for a new word. A new word judging means for judging a new word candidate character storing means for storing a character that is a candidate for a new word as a result of the judgment, and a new word for selecting a new word from the stored new word candidate characters in accordance with an instruction from the operator. An information storage / retrieval device comprising: a selection unit and a word dictionary registration unit that registers the selected new word as a known word in the word dictionary. 2. The information storage / retrieval device according to claim 1, wherein the new word selection means includes katakana character batch selection means for collectively selecting katakana characters from the stored new word candidate characters. 3. A word dictionary for storing known words, an extracting means for extracting words stored in the word dictionary from document information, and an index for searching the words for the extracted words. In a control method of an information storage and retrieval apparatus including an index assigning means, a set of characters other than the extracted words in the document information is targeted, and the characters belonging to the set are candidates for a new word. As a result of the determination, a character that is a candidate for a new word is stored in the new word candidate character storage means, a new word is selected from the stored new word candidate characters according to an instruction from the operator, and the selected new word is selected. A new word is registered in the word dictionary as a known word.