A Fuzzy Rule-Based Classification System Using Interval Type-2 Fuzzy Sets

摘要

The design of type-2 fuzzy rule-based classification systems from labeled data is considered in this study. With the aid of interval type-2 fuzzy sets, which can effectively capture uncertainties in the data, a compact and interpretable interval type-2 fuzzy rule base with fewer rules is constructed. Corresponding type-2 fuzzy reasoning method for classification is also presented. The validity of this classification system is shown through experimental results on several data sets.