Diabetes Classification using Fuzzy Logic and Adaptive Cuckoo Search Optimization Techniques

Abstract

Diabetic patients can be detected now a days globally. It�s main reason of growth is the incapability of body to produce enough insulin. So, majority of people today are either diabetic or pre-diabetic. Therefore, it is very much required to develop a system that can detect and classify the diabetes in optimal time period effectively and efficiently. So, proposed system make use of fuzzy logic and adaptive cuckoo search optimization algorithm (ACS) for diabetes classification. This work has been carried out in various steps. Firstly, the training dataset�s dimensionality reduction and optimal fuzzy rule generation via ACS optimization technique. Next is fuzzy model design and testing of fuzzified testing dataset. In this paper, outcome of FF-BAT algorithm has been compared with ACS algorithm. Experimental results were examined and it is noticed that ACS algorithm seems to perform better than FF-BAT algorithm.