Efficient and interpretable fuzzy classifiers from data with support vector learning

The maximization of the performance of the most if not all the fuzzy identification techniques is usually expressed in terms of the generalization performance of the derived neuro-fuzzy construction. Support Vector algorithms are adapted for the identification of a Support Vector Fuzzy Inference (SVFI) system that obtains robust generalization performance. However, these SVFI rules usually lack of interpretability. The accurate set of rules can be approximated with a simpler interpretable fuzzy system that can present insight to the more important aspects of the data. The interpretable fuzzy system construction algorithms receive an a priori description of a set of fuzzy sets that describe the linguistic aspects of the input variables as they are usually perceived by the human experts. In the case of the interpretable fuzzy sets an adaptive an algorithm for building them automatically is presented here. After the construction of the interpretable fuzzy partitions, the developed algorithms extract from the SVFI rules a small and concise set of interpretable rules. Finally, the Pseudo-Outer Product (POP) fuzzy rule selection orders the interpretable rules by using a Hebbian like evaluation in order to present the designer with the most capable rules. Key-Words: Interpretable Fuzzy Rules, Rule Mining, Support Vector Machines, Kernel Classifiers