Arrhythmia is caused due to the changes in the normal rhythm of heart with great risk of fatality if sustained over long periods of time. Many machine-learning algorithms have been developed for a fast diagnosis of cardiac arrhythmia. Also, the decision accuracies vary from method to method with Support Vector Machines (SVM) giving the highest. However, SVM cannot deal with the uncertainties in the process of diagnosis. Therefore, the authors have considered the Rough Set method for this classification. Analysis has been done with and without reducing the number of features. It has been found that the classification accuracies vary between 87% and 54% respectively. It may be emphasized that rough set reasons out all the possibilities and the decision could be as close as human diagnosis. The experimental results are presented. © Springer Science+Business Media B.V. 2008.
cited By (since 1996)0; Conference of org.apache.xalan.xsltc.dom.DOMAdapter@17810056 ; Conference Date: org.apache.xalan.xsltc.dom.DOMAdapter@163d1a28 Through org.apache.xalan.xsltc.dom.DOMAdapter@5804380; Conference Code:97249
M. Narwaria and Narayanankutty, K. A., “Classification of arrhythmia using rough sets”, in Innovations and Advanced Techniques in Systems, Computing Sciences and Software Engineering, Bridgeport, CT, 2008, pp. 326-329.