JP2804297B2 - Natural language processor - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a kana-kanji conversion processor used for inputting Japanese into a word processor or a DPS, a natural language analyzer for speech recognition, machine translation, proofreading support, character recognition, etc. It is also applied to a kanji conversion device and the like.

Conventional technologyConventionally, in a kana-kan conversion device that inputs a Japanese reading in kana, reads word data of kana / kanji from a word dictionary, and generates a sentence mixed with kanji kana, the word dictionary includes headings, notations, frequencies, Grammar information (part of speech etc.) was arranged. For this reason, when performing the kana-kanji conversion process, first, dictionary information corresponding to the headline was collected, and then grammatically dropped. However, as the number of dictionary searches for the kana-kanji conversion process increases, the processing amount of word data becomes enormous,
There is a disadvantage that the processing time is long. JP-A-59-112328 attempts to solve this problem by providing a two-stage index in the word dictionary. However, even in this method, if the dictionary information (for example, meaning / co-occurrence information) increases, the processing amount of the word data becomes enormous and the processing time becomes longer. In addition, unless the second index is referred to, the dictionary data of each word cannot be referred to, which is inconvenient.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described drawbacks. In a kana-kanji conversion process using a word dictionary, a natural dictionary system capable of reading necessary dictionary data at a high speed when necessary is provided. It is intended to provide a language processing device.

In order to achieve the above object, the present invention provides a kana-kanji conversion process for converting a kana character string into a character string mixed with kanji kana using a word dictionary that holds word information. It has a word dictionary provided with a plurality of pointers, and the pointer holds information to be minimized in each of a plurality of processes from the end of the input control process to before the output process. A pointer indicating an area in which each of the plurality of processes retrieves the heading, and reads dictionary information according to a pointer corresponding to each process. Hereinafter, a description will be given based on examples of the present invention.

FIG. 1 is a configuration diagram for explaining an embodiment of a natural language processing apparatus according to the present invention, wherein 1 is a keyboard, 2
Is an input control unit, 3 is a syntax analysis unit, 4 is a semantic analysis unit, 5 is a notation generation unit, 6 is an output control unit, 7 is a display, and 8 is a dictionary.

2 (a) to 2 (d) are diagrams showing a dictionary entry part.

In the case of the kana-kanji conversion process for performing a three-stage process on the input reading like the syntactic analysis unit 3, the semantic analysis unit 4, and the notation generation unit 5 to generate a kana-kanji mixed sentence,
As shown in the figure, pointers indicating areas holding the minimum necessary information in each of the above stages are stored in the dictionary headings. In the processing of the parsing unit, the dictionary heading (8-1) is searched from the input reading, and the word dictionary information is read and analyzed in accordance with the pointer (A) indicating the parsing information of the corresponding heading data. When performing the semantic analysis, the necessary minimum information such as the semantic classification is read by referring to the pointer (B) in the dictionary heading (8-1). At the time of notation generation, the pointer (C) is referred to.

Advantages As is apparent from the above description, according to the present invention, by holding a plurality of pointers pointing to dictionary information necessary for each processing in the heading part of the word dictionary, the necessary information can be obtained simply by referring to the heading part. Can be read, so that the amount of information handled at a time is reduced, and the processing time can be greatly reduced.

[Brief description of the drawings]

FIG. 1 is a configuration diagram for explaining an embodiment of a natural language processing device according to the present invention, and FIGS. 2 (a) to 2 (d) are diagrams showing a dictionary heading. DESCRIPTION OF SYMBOLS 1 ... Keyboard, 2 ... Input control part, 3 ... Syntax analysis part, 4 ... Semantic analysis part, 5 ... Notation generation part, 6 ... Output control part, 7 ... Display, 8 ... Dictionary.

──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Nobuyuki Oro 342 Minamikuma, Tottori-shi, Tottori Prefecture Ricoh Tottori Technology Development Co., Ltd. (56) References JP-A-61-33569 (JP, A) JP-A-60- 189571 (JP, A) JP-A-64-23769 (JP, A) JP-A-59-112328 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G06F 17 / 20-17 / 26

Claims (1)

(57) [Claims] 1. Using a word dictionary holding word information,
In the kana-kanji conversion process of converting a kana character string into a character string mixed with kanji and kana, the index has a word dictionary in which a plurality of pointers are provided for each heading in the index, and the pointer is output after the input control process is completed. A ponita indicating an area holding information to be minimized in each of a plurality of processes before the process, wherein each of the plurality of processes searches for the heading, A natural language processing apparatus for reading dictionary information in accordance with a pointer corresponding to a language.