Interval Type-2 Fuzzy Clustering Based Association Rule Mining Method

Searching interesting rules in transactional databases is considered as one of the most important data mining problems in many practical applications. Clustering based fuzzy association rule mining algorithms usually deal with data sets by clustering numerical values into boolean ones then using boolean method. However, the numerical data will have various disturbances in the practical application, as a result, the obtained association rules will change greatly. Therefore, the uncertainty in the association rule mining should be considered. In this paper, a type-2 fuzzy clustering based association rule mining algorithm (IT2FARM) is proposed, which mines the association rules among multi-dimensional data without prior knowledge. Meanwhile, we also design a method to test the number of correct fuzzy rules. Experimental results have shown that our proposed algorithm is superior in the number of association rules and the anti-noise capability compared with general clustering based methods.