Handling CHD Classifier Based on Machine Learning and Fuzzy Logic Techniques

Developing a high performance classifier requires understanding the complex relationships hidden within statistical information. Machine learning algorithms and rule based fuzzy logic approaches can help handle classification problems due to their efficient learning capabilities. Several approaches have been defined in both categories to classify discrete class labels. Selection of appropriate techniques and algorithms can help in developing high performance classifiers. In this paper, coronary heart disease (CHD) is taken as the classification problem whose objective is to find out whether an individual is suffering from CHD or not. To handle the CHD problem, four machine learning based classifiers and two fuzzy logic based classifiers have been developed. Accuracy, precision, recall and Fl-score evaluation metrics have been used to evaluate the performance of the developed classifier. The results obtained from the developed classifiers showed that Support Vector Machine and Wang-Mendel based classifiers perform with 92.32% and 98.48% accuracy in their respective category.