Soft Set Approach for Maximal Association Rules Mining

In this paper, an alternative approach for maximal association rules mining from a transactional database using soft set theory is proposed. The first step of the proposed approach is based on representing a transactional database as a soft set. Based on the soft set, the notion of items co-occurrence in a transaction can be defined. The definitions of soft maximal association rules, maximal support and maximal confidence are presented using the concept of items co-occurrence. It is shown that by using soft set theory, maximal rules discovered are identical and faster as compared to traditional maximal and rough maximal association rules approaches.