Hierarchical Fuzzy rule based classification systems with entropy-based rule weighting for imbalanced data sets

In the field of classification, many real world application are imbalanced. It mean that the number of instances from some classes are much higher than that of the other classes. Most of classification techniques tend to classify majority classes correctly and therefore many instances of minority classes misclassified. In this study, we propose a Fuzzy Rule Based Classification (FRBC) system using a hierarchical rule learning method in the case of imbalanced data-sets. In each stage of the hierarchy, a set of rules with certain length of antecedent are investigated. A novel rule weighting method, based on the entropy measure, determines the appropriateness of each rule. The effectiveness of the proposed method is completed with statistical analysis over imbalanced data sets especially in tackling the tradeoff between accuracy and comprehensibility of fuzzy rule-based systems.