Closed set based discovery of representative association rules

Abstract

Discovering association rules among items in a large database is an important database mining problem. However, the number of association rules may be huge. The problem can be alleviated by applying concise lossless representations of association rules. There were proposed a few such representations in the late ninetieths. Representative association rules are such an example representation. The association rules, which are not representative ones, may be derived syntactically from representative rules by means of a cover operator. In the paper we show how to discover all representative rules using only closed itemsets and their generators.