An algorithm for rule selection on fuzzy rule-based systems applied to the treatment of diabetics and detection of fraud in electronic payment

Recently, many papers have proposed automatic techniques for · extraction of knowledge from numerical data. · minimization of the number of rules. But few works have been developed for design of experiments and datum plane covering. Most of optimization methods make the assumption that datum plane is sufficiently covered. If this assumption no longer holds, we will see that these methods may not work, since it implies that, before optimization, the fuzzy system gives acceptable results. We present in this paper, an algorithm for selection of fuzzy rules based on datum plane covering. We apply this method to two applications: · the problem of treatment of diabetics. Taking the insulin infusion rate as the input and the blood glucose rate as the output, we consider the patient as a black box, whose model has to be obtained from the available measures of inputs-outputs. We dispose of a glycaemia file automatically produced for every person, and an insulin file shared by several persons. · the problem of detecting fraud in electronic payment systems. Along with the development of credit cards, fraudulent activities become a major problem in electronic payment systems. Our model is based on specific and usual customer behavior, and deviation from such patterns is suspect.

I. Introduction

Many works have proposed a minimization of the number of rules [2], [3], [4]. Ishibushi in [2] defines candidate rules based on the grade of certainty and applies genetic algorithm for rule selection problem. Alcala and co, in [3], estimate zones with bad rules, with redundant or irrelevant rules and complementary rules, and search a good configuration of rules by removing rules with little importance.