Mining relaxed closed subspace clusters

This paper defines and discusses a new problem in the area of subspace clustering. It defines the problem of mining closed subspace clusters. This new concept allows for the culling of more high quality and less redundant clusters, than that of traditional clustering algorithms. In addition, our method contains a relaxation parameter, which allows for the classification of qualifying clusters into mutually exclusive bins of varying quality--extending the problem to mining relaxed closed subspace clusters. These concepts culminate in a new algorithm called Relaxed Closed Subspace Clustering (RCSC). Categories and Subject Descriptors H.2.8 [Database Management]: Database Applications-data mining; H.3.3 [Information Storage and Retrieval]: Information Search and Retrieval--Clustering General Terms Algorithms, Theory. Keywords Closed subspace cluster, maximal subspace cluster, relaxed closed subspace cluster, relaxed interval, subspace clustering, data mining.

• Type: Journal
• Year: 2010
• Where: ACMSE
• Authors: Erich Allen Peterson, Peiyi Tang
• Keywords: ACMSE 2010, Maximal Subspace Cluster, Mining, Subspace Cluster, Theoretical Computer Science