Using concept lattices for disjoint clustering
Abstract
Clustering is a well studied problem in data mining. Most clustering research has focused on generating disjoint clusters with a recently increasing interest in overlapping clustering [1, 2]. Concept lattices from Formal Concept Analysis provide a convenient and relatively easy way for over-lapping clustering. On the other hand, disjoint clustering with concept lattices is an open problem. In this paper, we present a solution to this open problem. Our solution uses frequent itemsets from association rule mining.
Department(s)
Computer Science
Document Type
Conference Proceeding
Keywords
Clustering, Concept Lattice, Data and Text Mining, Formal Concept Analysis, Frequent Itemsets, Machine Learning
Publication Date
12-1-2003
Recommended Citation
Saquer, Jamil M. "Using concept lattices for disjoint clustering." Proceedings of Information Knowledge and Sharing IKS, Scottsdale, AZ, USA (2003).
Journal Title
Proceedings of the IASTED International Conference on Information and Knowledge Sharing
