Mining Valid Association Rules in Incomplete Information Systems

Extracting the association rules is an important research topic among the various data mining problems. Based on the rough set theory which is a powerful tool in dealing with incompleteness and uncertainty, an algorithm to mine association rules in incomplete information systems is presented. In the new algorithm, the support and confidence are redefined and a new judgment criterion is introduced in. The algorithm can mine the positive, invalid and negative association rules directly without processing missing values. The experiment shows that the new algorithm has short execution times, and can mine effective association rules efficiently.