| 摘要 |
PROBLEM TO BE SOLVED: To reduce a burden on a user by automating the identification of a text kind and the selection of a pre-editing rule and standard notation corresponding to the kind to rewrite in the pre-editing processing of a text described in natural language. SOLUTION: When an input part 2 inputs a text described in natural language, the morpheme analytic part 5a of a program memory 5 morpheme- analyzes each word of the inputted text by referring to a dictionary table 4a in a table memory 4, and extracts part-of-speech information and morpheme information. Next, an identification part 5b identifies the kind of the text from extracted part-of-speech information and morpheme information by referring to an identification rule table 4b. Then a pre-editing part 5c detects a word at the pre-editing object part of the text from an editing rule corresponding to the identified kind by referring to an editing rule table 4c, and writes it to standard notation. Thus, selecting processing of a group of pre-editing rules according to various kinds of text kinds is automated to reduce the burden on the user. |
