作者: Yves Bastide , Nicolas Pasquier , Rafik Taouil , Lotfi Lakhal
DOI:
关键词: Association rule learning 、 Set (abstract data type) 、 Generating set of a group 、 Computer science 、 Data mining 、 Limit (mathematics) 、 Closed set
摘要: In this paper, we address the problem of understandability and usefulness set discovered association rules. This is important since real-life databases lead most time to several thousands rules with high confidence. We thus propose new algorithms based on Galois closed sets limit extraction small informative covers for exact approximate rules, structural Once frequent itemsets - which constitute a generating both have been discovered, no additional database pass needed derive these covers. Experiments conducted show that are efficient valuable in practice.