Абстрактный

Discovering Multi-Level Association Rules using Fuzzy Hierarchies

Usha Rani, R. Vijaya Prakash, A. Govardhan

In this paper, Fuzzy concept hierarchies are used for multi-level association rule mining from large datasets via Attribute-Oriented Induction approach [1]. In this the process of fuzzy hierarchical induction approach is used and extends it with two new characteristics which improve applicability of the original approach in data mining. The proposed drilling-down approach of fuzzy induction model allows user to retrieve estimated explanations of the generated abstract concept. An application to discovery of multi-level association rules from environmental data stored in a Toxic Release Inventory is presented.