|Budget Amount *help
¥3,900,000 (Direct Cost : ¥3,900,000)
Fiscal Year 1993 : ¥900,000 (Direct Cost : ¥900,000)
Fiscal Year 1992 : ¥1,000,000 (Direct Cost : ¥1,000,000)
Fiscal Year 1991 : ¥2,000,000 (Direct Cost : ¥2,000,000)
The prpose of this research is to fully formalize the Japanese gramr for computer processing of Japanese. We have used a great number of essays, novels scientific papers and so on, which were actually written, as data, and analized and described rules of word and sentence structures of Japanese in accordance with the following procedure. 1. definition of the word, 2. classification of words, 3. analysis and description of positions inside words and co-occurrance of morhemes words, 4. description of surface sentence patterns in terms of co-occurrence of words, 5. description of case frames in terms of semantic ralations, 6. adjustment of word formation rules, 7. adjustment of sentence structure rules, 8. arrangement of computer dictionary of Japanese, and 9. production of algorisms and computer programs of morphological and syntactic analyzes.
Since syntactic rules of sentences which were actually written are so complicated and we are still in working for rule description, we intend to publisize formalization and computer programs of syntactic rules shortly.
The mail feature of this research is that differently from usual definitition of Japanese word, we define it in terms of simultaneity and indivisibilty, so that morphological analyzes which can use regular rules and syntactic analyzes which use context free or sensitive rules can be separated completely.