Abstract

Studying how to make the results have more guiding significance is an important subject of association rules mining. It is an effective solution to screen and sort mining result according to the user's prior knowledge and subjective interestingness measure. This paper puts forward IAFBRS algorithm, brings in rule schema to describe the knowledge the user owned, shows users' subjective interestingness measure for rules through the rule schema, and does different treatment with rules of different interestingness measure, then uses LRU algorithm to optimize the rule schema level and the mining rules. At last, this paper shows the superiority of IAFBRS algorithm for mining association rules through experiments.

Paper Details