ラフ集合論に基づく知識指向型クラスタリング法(<一般講演>ラフ集合・分析,医療情報システムにおけるソフトコンピューティグ)

抄録

This paper presents a knowledge-oriented clustering method based on rough set theory. The method evaluates simplicity of classification knowledge during the clustering process and produces readable clusters reflecting global feature of objects. The method uses a newly introduced measure, indiscernibility degree, to evaluate importance of equivalence relations that is related to roughness of the classification knowledge. Indiscernibility degree is defined as a ratio of equivalence relations that give common classification to two objects under consideration. The two objects can be classified into the same class if they have high indiscernibility degree, even in presence of equivalence relations which differentiate these objects. Ignorance of such equivalence relations is related to generalization of knowledge, and yields simple clusters that can be represented by simple knowledge. An experiment was performed on the artificially created numerical datasets. The results showed that objects were classified into the expected clusters if modification was performed, whereas they were classified into many small categories without modification.