Finding Intensity of Disease by Fuzzy Learning Rules (FLR) based Intelligent Decisive Support System for Efficient Disease Prediction

Abstract

Experimental analysis is frequently done by doctor‟s expertise and involvement. But still, cases are testified of incorrect diagnosis and action. Patients are demanded to take some tests for examination. In many belongings, not all the tests contribute towards efficient analysis of a disease. Three classifiers similar Naive Bayes, Ordering by clustering and decision tree are used to calculate the analysis of patients with the same accuracy as acquired before the discount of some attributes. Fuzzy learning rules (FLR) are generally applied for finding the intensity of disease in data sets. Fuzzy learning rules better compared to other three methods. We propose a fuzzy learning algorithm to determine relationships between data resources based on their disease attributes, as well as to characterize knowledge through the connotation of disease covered by those properties. The algorithm addresses the significant problem of important a suitable number of clusters for suitably catching all the diseases of the knowledge domain. Using fuzzy rule-based classification system, the proposed system proves to improve the classification accuracy.