JP2592993B2 - Phrase extraction device - Google Patents

Description

The present invention relates to a phrase extracting apparatus, and more particularly, to a phrase extracting apparatus suitable for a language processing device such as a word processor, a translation device, a proofreading device, and a device using a database. The present invention relates to a phrase extracting device.

(B) Conventional technology Conventionally, in order to extract a phrase for translation or proofreading from a converted Japanese document into which the information of the phrase delimiter is not inserted, an independent word dictionary, an auxiliary word dictionary, an affix dictionary, etc. Dictionaries and a table showing the connection relationship of those elements,
The segmentation was performed while referring to tables such as a grammar table.

(C) Problems to be Solved by the Invention In such a conventional phrase extracting apparatus using dictionaries and tables, (1) a large amount of storage devices for storing the dictionaries and tables are required. Become.
(2) The time required for the phrase extraction processing for performing a dictionary search or a table search becomes longer. (3) There is a problem that a control program for segment extraction becomes complicated. Also,
In phrase extraction processing, a device that extracts a phrase by determining the character type of each character in a character string has also been considered. However, in such a device, since a phrase is mechanically cut out according to the character type, when there is a character including a kana notation prefix, there is a case where a wrong phrase is cut out to separate the prefix from the character.

The present invention has been made in view of the above circumstances, and provides a phrase extracting apparatus that can solve the above problem.

(D) Means for Solving the Problems FIG. 1 is a block diagram showing the basic configuration of the present invention. Referring to FIG. 1, the present invention provides a sentence storing means 1 for storing a sentence, and an instructing means 2 for reading out a desired sentence from the sentence storing means 1 in sentence units and instructing to extract a sentence.
Reading means 3 for reading a designated sentence from the sentence storage means 1 in sentence units including a period, a character type and a kanji character of a hiragana character one by one based on the character code of the read sentence,
The character type and punctuation of non-Hiragana characters such as katakana are converted by the character type determining means 4 and the Hiragana character and the non-Hiragana character and punctuation determined by the character type determining means 4 are respectively associated with two types of codes. Conversion means 5 and determination result storage means 6 for sequentially storing two types of code strings of the converted text
And a displacement point of the code string when changing from a Hiragana character to a non-Hiragana character is detected for the two types of code strings stored in the determination result storage means 6, and a break in a sentence corresponding to the detected displacement point. A phrase extracting means 7 for inserting
Prefix correction dictionary means 8 for storing a phrase break error pattern and its correction pattern for a character string including a kana notation prefix, and a phrase break error pattern in a sentence in which a phrase break is inserted It is checked whether or not the error pattern exists, and if the error pattern exists, the arrangement of the codes of the displacement points of the corresponding code string in the determination result accumulation means 6 is changed based on the correction pattern, thereby changing the break of the phrase. A phrase extraction and correction means 9 for extracting a phrase again, a storage means 10 for storing a sentence including a break of the phrase changed by the phrase extraction and correction means 9, a sentence stored in the storage means 10 and a break of the phrase This is a phrase segmentation device including output means 11 for visually outputting.

Preferably, the two kinds of codes are constituted by binary digital signals.

In the present invention, it is preferable that the instruction means 2 is constituted by an input device including a keyboard, a mouse, and the like.

Sentence storing means 1, reading means 3, character type determining means 4, converting means 5, determining result storing means 6, phrase extracting means 7, prefix correction dictionary means 8, phrase extracting correcting means 9, storage means 10
Is preferably configured by a microcomputer comprising a CPU, a ROM, a RAM, and an I / O port, and an ASIC. In particular, the text storage means 1 and the storage means 10 are constituted by the RAM therein,
Preferably, the prefix correction dictionary means 8 is constituted by a ROM.

The output unit 11 is preferably configured by a display device including an LCD (liquid crystal) display, a CRT display, and the like.

The phrase segmentation device according to the present invention can be applied to a word processor, a translation device, a proofreading device, a device that uses a database, and the like. In addition, a phrase segmentation process is required even in a device that outputs sentences as speech. Can be applied.

(E) Function According to the present invention, when a desired sentence is read out from the sentences stored in the sentence storage means 1 in units of sentences, the character type and the kanji of the hiragana character are determined one by one based on the character code of the read sentence. The character type and punctuation of non-Hiragana characters, such as katakana, are distinguished by the character type discriminating means 4, and the determined Hiragana character and the non-Hiragana character and punctuation are converted by the converting means 5 to correspond to the two kinds of codes, respectively. And
The two types of code strings of the converted text are stored in the determination result storage unit 6.
Are sequentially accumulated. Next, the phrase extracting means 7 detects a displacement point of the code string when the hiragana character changes from the hiragana character to the non-hiragana character in the two kinds of code strings stored in the discrimination result storage means 6, and a sentence corresponding to the detected displacement point. Insert a break between phrases. The phrase cut-out correcting means 9 refers to the prefix correction dictionary means 8 storing the error pattern of the phrase break and the correction pattern for the character string including the kana notation prefix, and inserts the phrase break. Check if there is an error pattern of segment breaks in the sentence,
When the error pattern exists, the arrangement of the codes of the displacement points of the corresponding code string of the discrimination result accumulating means 6 is changed based on the correction pattern, thereby changing the break of the phrase and again extracting the phrase. . The sentence including the changed segment break is stored in the storage means 10 and further stored in the storage means 10
The output means 11 serves to visually output the sentence stored in the file and the break of the sentence.

Therefore, when the segment extraction is performed, the segment extraction process is performed by reading out the sentence unit including the period, thereby eliminating the complicated segment extraction operation.

Furthermore, in the segment extraction processing, based on the character code of the sentence read in the sentence unit, a hiragana character, a non-hiragana character such as a kanji character, a katakana character, and a punctuation mark are discriminated one by one and converted into two kinds of code strings. Since the phrase extraction is mechanically performed at the displacement point of the code string when converting from Hiragana characters to non-Hiragana characters, dictionaries for phrase extraction are unnecessary, and phrase extraction is performed at high speed for each sentence unit. Can be. Thereby, the configuration of the control program and the device can be simplified, and the IC can be implemented.

In addition, even if there is a kana notation prefix in the sentence, the configuration is provided with a prefix correction dictionary means for automatically correcting an error in segmentation due to the kana notation prefix. it can.

(F) Embodiment The present invention will be described in detail below based on an embodiment shown in the drawings.
The present invention is not limited by this.

FIG. 2 is a block diagram showing an embodiment in which the present invention is applied to a word processor. In the figure, reference numeral 20 denotes a word processor. Reference numeral 21 denotes a text storage device which can be composed of, for example, a floppy disk or hard disk as an external storage device, for example, a RAM as an internal storage device, or a database as another storage device, and is composed of kana-kanji mixed sentences. Japanese documents are accumulated. Reference numeral 22 denotes a keyboard as instruction means, which is provided with character input keys for performing text editing, text correction, etc., various instruction keys, etc., for inputting text and reading a desired text from the text storage device 21. Enter instructions for Reference numeral 23 denotes a reading device that cooperates with the CPU 24, and reads a sentence specified by the keyboard 22 from the text storage device 21. Numeral 25 is a character type discriminating device which cooperates with the CPU 24. The character type such as kanji, hiragana, katakana, etc. is discriminated one character at a time in the sentence read out from the sentence storage device 21, and the punctuation mark is discriminated. Numeral 26 denotes a result accumulating device as a judgment result accumulating means and a memory means. The result accumulating device is constituted by a RAM, and sequentially accumulates the character type judgment results, and stores a sentence cut by a later-described phrase cutout device.

Numeral 27 is a character type digitizing device as a conversion means cooperating with the CPU 24. The first code, specifically "H", is used for the kanji and katakana as the character type discrimination result by the character type discriminating device 25. ", And a second code, specifically" L ", for the hiragana and the punctuation mark, thereby converting the character type discrimination result to one of the two codes" H "or" L ". replace. And the codes “H” and “L” which are the result of character type determination
Are stored in the result storage device 26 via the CPU 24.

The phrase extraction device 28 cooperates with the CPU 24 to
When a predetermined code transition point is detected for the code strings “H” and “L” stored in 6, a break in the sentence is inserted in the read sentence.

Reference numeral 29 denotes a kana prefix correction table which stores an error pattern of a segment break and its correction pattern for a character including the kana prefix.

Numeral 30 is a kana notation prefix processing device as a kana notation prefix correction means, which checks whether there is an error in the kana notation prefix in the sentence into which the previous knot is inserted, and checks the kana notation error. If there exists, the arrangement of the code strings in the result storage device 26 is changed, thereby changing the break of the phrase, and extracting the phrase again.

Reference numeral 31 denotes a display device as an output unit connected to the CPU 24 via the output control unit 32, which is configured by a dot matrix type display device such as a CRT or an LCD, so that a break in a phrase can be confirmed. Displays the sentence where the break was inserted.

In such a configuration, the phrase segmentation process will be described according to the example sentence "We have prepared a material to be sent to the JPO and will contact you." In accordance with the flowchart shown in FIG.

FIG. 3 shows a state in which, for example, one sentence is cut out from the sentence storage device 21 in which a sentence file and other databases are stored in accordance with the processing unit, and the sentence is stored in the result storage device 26. As described above, when a desired sentence is read from the sentence storage device 21 in accordance with the processing unit, for example, one sentence, one paragraph, one chapter, and the like (step 40), the character type constituting the sentence is read. The code is determined (step 41). Specifically, since each character of the read sentence is assigned a unique character code such as a JIS code, by comparing the character code with each condition of the code determination table shown in FIG. , To determine the character type. Cc in condition
Are the characters to be used for character type determination, a1 and b1 represent the beginning and end of the Kanji code, a2 and b2 represent the beginning and end of the Hiragana code, a3 and b3 represent the beginning and end of the Katakana code, and a4 Represents a period.

The determined character type code is stored in the result storage device 26 (step 42) and digitized by the character type digitizing device 27 (step 43). FIG. 5 shows a code digitization table referred to for digitizing the determined output code. That is, if the character type is determined to be kanji, it is converted to the code "H", and if it is determined to be hiragana, it is converted to the code "L".
, The phrase is replaced by the code "L" respectively. The result of digitizing the sentence by collation with the code digitization table is stored in the result storage device 26 as a code string shown in FIG. In FIG. 6, example sentences are also shown for explanation.

Next, the digitized result is compared with a cut-out determination table shown in FIG. 7 to determine a segment break (step
44). The determination of a segment break is as follows: (1) A break of a keyword (phrase) is made at a transition point between “L” and “H” in the digital output. (2) Discrimination is made by inserting a break between keywords after a period.

Next, a symbol, for example, "/" cut out at the break based on the determination of the break by collation with the cut-out determination table is inserted, and the sentence into which the cut-out symbol is inserted is stored in the result storage device 26 (step 45). ). FIG. 8 shows the phrase segmentation result obtained by the above processing. In the figure, the phrase is correctly extracted according to the phrase extraction determination table for characters that do not include the kana notation prefix, but the phrase extraction is incorrect for the phrase that includes the kana notation prefix "go". I understand.

Kana notation prefixes include "o" and "go", which are used relatively frequently in the text. Also, "O",
"Go" has the property of being easily associated with Japanese and Chinese.

The kana spelling prefix correction table 29 of this embodiment is provided to correctly extract a phrase even if a kana spelling prefix exists, and the kana spelling prefix error pattern “O / Kan”, “Go / Kan”, their digitized code strings “L / H”, and their modified digitized code strings “/ HH” are stored.

In order to eliminate a kana notation prefix extraction error, a segment break is first searched for using the extraction result shown in FIG. Next, the phrase break error pattern shown in FIG. 9 is compared with the phrase before and after the phrase in the character type discrimination result shown in FIG. 6 (step 46), and it is checked whether the phrase break matches the error pattern (step 46). Step 47).

If there is a part that matches the error pattern in the character type discrimination result, the arrangement pattern composed of the digitized code and the phrase cutout symbol is corrected according to the kana notation prefix correction table 29 (steps 48 to 49). The corrected results are shown in FIG. 10 and FIG.

Next, if the end condition is no, that is, if there is a next sentence to be segmented, control is performed to read the next sentence from the sentence storage device 21 and to control that the storage positions do not overlap (step 50). ).

If yes in step 50, that is, if there is no longer any object to be cut out, the necessary information is stored in the result storage device 26.
And the process ends (step 51).

(G) Effects of the Invention According to the present invention, when a segment is extracted, a segment extraction process is performed by reading out a sentence unit including a period, thereby eliminating a complicated segment extraction operation.

Furthermore, in the segment extraction processing, based on the character code of the sentence read in the sentence unit, a hiragana character, a non-hiragana character such as a kanji character, a katakana character, and a punctuation mark are discriminated one by one and converted into two kinds of code strings. Since the phrase extraction is mechanically performed at the displacement point of the code string when changing from a Hiragana character to a non-Hiragana character, dictionaries for phrase extraction are unnecessary, and phrase extraction is performed at high speed for each sentence unit. Can be. Thereby, the configuration of the control program and the device can be simplified, and the IC can be implemented.

Also, even if there is a kana notation prefix in the sentence, the configuration is provided with a prefix correction dictionary means for automatically correcting the error of segmentation due to it, so that the erroneous segment extraction process is prevented. Can be.

Therefore, the present invention can be applied not only to word processors and office computers, but also to other small information devices, specifically, electronic notebooks and calculators with program functions.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a basic configuration of the present invention, FIG. 2 is a block diagram showing a configuration of a word processor according to an embodiment of the present invention, and FIG. 3 is an example of a sentence stored in a text storage device. FIG. 4 is an explanatory diagram showing the contents of a character type discrimination table, FIG. 5 is an explanatory diagram showing the contents of a code digitization table, FIG. 6 is an explanatory diagram showing the character type discrimination results, and FIG. Is an explanatory diagram showing the contents of the phrase extraction determination table, FIG. 8 is an explanatory diagram showing the extraction result, FIG. 9 is an explanatory diagram showing the contents of the kana notation prefix correction table, and FIG. 10 is an explanatory diagram showing the correction result FIG. 11 is an explanatory diagram showing an output state of a correction result, and FIG. 12 is a flowchart showing a phrase extracting process. 1 ... sentence storage means, 2 ... instruction means, 3 ... readout means, 4 ... character type determination means, 5 ... conversion means, 6 ... determination result storage means, 7 ... phrase extraction means, 8 ... ... Prefix correction dictionary means, 9 ... Phrase segmentation correction means, 10 ... Storage means, 11 ... Output means.

Claims (1)

(57) [Claims] A sentence storing means for storing a sentence; an instructing means for reading out a desired sentence from the sentence accumulating means in units of sentences and instructing a segment cutout; and a sentence including a period from the sentence storing means. Reading means for reading in units; character type discriminating means for discriminating the character type of hiragana characters and the character type and punctuation of non-hiragana characters such as kanji and katakana character by character based on the character code of the read text; Conversion means for converting the hiragana character, the non-hiragana character, and the punctuation mark determined in accordance with the two types of codes, respectively, and a determination result storage means for sequentially storing two types of code strings of the converted text. A displacement point of the code string when changing from a Hiragana character to a non-Hiragana character is detected for the two types of code strings accumulated in the result accumulating means, and corresponds to the detected displacement point. In the sentence, a phrase break is inserted into the phrase, a prefix correction dictionary storing the error pattern of the phrase break and its correction pattern for the character string including the kana notation prefix, and a phrase break is inserted. It is checked whether there is an error pattern of a segment break in the sentence, and if the error pattern exists, the arrangement of the codes of the displacement points of the corresponding code string in the determination result storage means is changed based on the correction pattern. ,
Thus, a phrase cut-out correcting unit that changes the break of the phrase and cuts out the phrase again, a storage unit that stores a sentence including the break of the phrase changed by the phrase cut-out correcting unit, and a sentence stored in the storage unit. A phrase segmentation device comprising output means for visually outputting the segment break.