METHOD FOR GROUPING SET OF WORDS CAPABLE OF APPEARING IN SET OF NATURAL LANGUAGES

摘要

PROBLEM TO BE SOLVED: To exactly and speedily group words in natural languages by automatically clustering words in the set of words while using the relation value of suffix relation data, and providing a suffix in the group and group data having equivalent character strings before the suffix. SOLUTION: The input of a box 10 provides information concerning the natural language set and the operation of a box 12 provides the suffix relation data out of that information into a box 14 as the intermediate result. The suffix relation data in the box 14 and the set of words (word 1 to word M) are used for the operation of a box 22. In this operation, the words are clustered while using the relation values and as the final result, the group data showing word groups (word group 1 to word group N) are provided in a box 24. The word group has the suffix similar to one of relations and two or more words having the equivalent character strings before the suffix.