Fuzzy association rules in soft conceptual hierarchies
Trevor Martin, Yun Cheng Shen · 2009
Humans frequently use a ldquodivide and conquerrdquo strategy to understand large volumes of data, by grouping similar items into progressively finer categories which form a conceptual hierarchy. Typically, such categories do not have crisp definitions but can be modelled by fuzzy set theory, allowing computers to represent and reason about sets of objects in a way that reflects the human interpretation of categories. Association rules are a useful tool in knowledge discovery from databases but are normally defined in terms of crisp rather than fuzzy categories. In this paper, we describe a new approach to finding association rules between fuzzy categories, based on mass assignment theory. In contrast to other fuzzy association methods, we retain a fuzzy confidence value rather than a point value.