摘要 |
PURPOSE:To estimate the type of the word conjugation from the declension and the information on parts of speech by using a dictionary updating means that has a function to estimate the information on the dictionary used for natural language processing in parallel with the natural language processing. CONSTITUTION:In case a word is registered to a dictionary of a Japanese word processing system, a display is given to supply the reading and the KANJI (Chinese characters) description of the word to be registered in the end-form expression if the word has the conjugation or in the participial adjective expression that modifies after the substantives if no end-form is available. Then the KANA rendering character string and the KANJI character string are inputted for the word to be registered and inputted through a keyboard 7 together with its part-of-speech number. For instance, the processes related to a verb, an adjective and an adjective verb are carried out with the part-of-speech numbers '4', '5' and '6' respectively. It is regarded that other numbers mean the words having no conjugation. Then the data record where the information on the rendering KANA character strings, KANJI character strings, parts-of-speech numbers, the conjugation types, etc., are arranged according to a format is stored in a prescribed position of a dictionary 8. In such a way, the effective update is secured for the dictionary.
|