Mining Direct and Indirect Fuzzy Multiple Level Sequential Patterns in Large Transaction Databases

Abstract

Sequential pattern is an important research topic in data mining and knowledge discovery. Traditional algorithms for mining sequential patterns are built on the binary attributes databases, which has three limitations. The first, it can not concern quantitative attributes; the second, only direct sequential patterns are discovered; the third, it can not process these data items with multiple level concepts. Mining fuzzy sequential patterns has been proposed to address the first limitation. We put forward a discovery algorithm for mining indirect multiple level sequential patterns to deal with the second and the third limitations, and a discovery algorithm for mining both direct and indirect fuzzy multiple level sequential patterns by combining these three approaches.