ECG Arrhythmia Classification Using Spearman Rank Correlation and Support Vector Machine

This paper presents a diagnostic system for the detection of presence and absence of cardiac Arrhythmia from the Electrocardiogram (ECG) data using the methods of Feature Selection, Feature Extraction and Binary Classification Technique. A hybrid approach of three algorithms namely Rank Correlation, Principal Component Analysis (PCA) and Support Vector Machine (SVM) are applied on the UCI Cardiac Arrhythmia data set for the automatic arrhythmia detection in Arrhythmia Diagnostic System. Spearman Rank Correlation aids the process of dimension reduction and increases the accuracy of the classifier. In this study, SVM has been widely used for classification based diagnosis of diseases. The results obtained after implementation of all the three methods are thus compared as per their accuracy rate in percentages and the performance of the SVM classifier depends on the cost and kernel parameter sigma classification frequency upon the number of attributes selected by Rank Correlation. The experimental method shows that hybrid approach is superior to other approaches.