Weighted Fuzzy Rule-Based System Combined With A Novel Simplified E.Coli Foraging Optimization Algorithm

A simplified E.Coli foraging optimization algorithm is presented in this paper, which simulates the chemo-tactic behavior of E.Coli. The optimization algorithm characterizes the easy implementation and the fact that no gradient information is required. The operators of the algorithm are described in details. The simplified E.Coli algorithm consists of a tumbling operator and a swimming operator. At the same time the optimal position of individual E.Coli and the location of all E.Coli swarm are adopted to update the locations of swarm. In this paper, a weighted fuzzy rule-based system has been designed, in which the parameters of membership functions including position and shape of the fuzzy rule set and weights of rules are estimated using the simplified E.Coli foraging optimization algorithm. The efficiency of the system has been illustrated in the process of classifying the iris data. This paper also shows that weighted fuzzy rules can lead to better fuzzy system compared with non-weighted fuzzy rules.

I. INTRODUCTION

Since L.A. Zadeh first introduced Fuzzy logic in 1965, it has emerged as a powerful technique for the industrial control process, diagnosis and other rule-based systems. Fuzzy systems can be considered as knowledge-based systems, incorporating human knowledge into their Knowledge Base through Fuzzy Rules and Fuzzy Membership Functions. The definition of these Fuzzy Rules and Fuzzy Membership functions is generally affected by subjective decision, which greatly influences the system performance. Therefore, many optimization methods have been proposed to decide the related parameters of Fuzzy Membership functions or Fuzzy Rules[1] [2] [3] [4] [5]. Genetic algorithms (GAs) have been applied to learn both the antecedent and consequent part of fuzzy rules and models with both fixed and varying number of rules [1] [2]. All possible rules, position and type of membership function are learned by means of GA [3] [4]. The parameters of fuzzy system are obtained using knowledge bounded least squares method [5].