On hybridizing fuzzy min max neural network and firefly algorithm for automated heart disease diagnosis

• Published in: 2013 Fourth International Conference on Computing, Communications and Networking Technologies (ICCCNT) pp. 1 - 5
• Main Authors: Rajakumar, B. R., George, Aloysius
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.07.2013
• Subjects: Accuracy; Classification; Classification algorithms; Conferences; Diseases; Firefly Algorithm; Fuzzy Min-Max Neural Network (FMMN); Heart; Heart Diseases; IEEE conference proceedings; Neural networks; Open Hyperbox; Rule Mining
• DOI: 10.1109/ICCCNT.2013.6726611

Heart disease is the most important reason of morbidity and mortality in the modern society. For that reason, it is important to have a proper diagnosis of heart disease for patients to live to tell the tale. In order to make the diagnosis system as the efficient one, heart diseases should be classified accurately. In the existing technique, the quality of the extracted rules is poor. So as to increase the quality of the extracted rules, an efficient technique should be used. In our proposed methodology, we are using firefly algorithm in Fuzzy Min-Max Neural Network. Firefly algorithm has high convergence tempo. It works individually and finds a superior position for itself in contemplation with its recent position as well as the situation of other fireflies. And it escapes from the local optima and finds a global optimum which has a smaller amount number of iterations. Since it is a robust algorithm, the classification of heart diseases can be done fastly and as a result the accuracy and performance of the proposed technique becomes encouraging.