Clustering association rules
Brian LentDept. of Comput. Sci., Stanford Univ., CA, USAA. SwamiDepartment of Computer Science, University of Stanford, Stanford, CA, USAJennifer WidomDepartment of Computer Science, University of Stanford, Stanford, CA, USA
2002en
ABI
Abstract
The authors consider the problem of clustering two-dimensional association rules in large databases. They present a geometric-based algorithm, BitOp, for performing the clustering, embedded within an association rule clustering system, ARCS. Association rule clustering is useful when the user desires to segment the data. They measure the quality of the segmentation generated by ARCS using the minimum description length (MDL) principle of encoding the clusters on several databases including noise and errors. Scale-up experiments show that ARCS, using the BitOp algorithm, scales linearly with the amount of data.
Identifiers
Citations and references
Cited by 30 references