IMPROVED FUZZY ARTIFICIAL NEURAL NETWORK (IFANN) CLASSIFIER FOR CORONARY ARTERY HEART DISEASE PREDICTION IN DIABETES PATIENTS

B. Narasimhan; Dr. A. Malathi

Volume : IX, Issue : IV, April - 2019

IMPROVED FUZZY ARTIFICIAL NEURAL NETWORK (IFANN) CLASSIFIER FOR CORONARY ARTERY HEART DISEASE PREDICTION IN DIABETES PATIENTS

B. Narasimhan, Dr. A. Malathi

Abstract :

Soft computing techniques and its applications extends its wings in almost all areas which includes data mining, pattern discovery, industrial applications, robotics, automation and many more. Soft computing comprises of the core components such as fuzzy logic, genetic algorithm, artificial neural networks and probabilistic reasoning. In spite of these, recently many bio – inspired computing attracted attention for the researchers to work in that area. Machine learning plays an important role in the design and development of decision support systems, applied soft computing and expert systems applications. This research work aims to build an improved fuzzy logic based artificial neural network classifier for predicting coronary artery heart disease among diabetic patients. Real time data are obtained and the built IFANN classifier is compared with Takagi Sugeno Kang fuzzy classifier and ANN classifier in terms of prediction accuracy, sensitivity, specificity and Mathew’s correlation coefficient. The significance of MCC is that to test the ability of the machine learning classifier in spite of other performance metrics. Implementations are done in Scilab and from the obtained results it is inferred that the built IFANN outperforms that that of TSK fuzzy classifier and ANN classifier.