Weighted SVM with RR Interval based Features for Android-based Arrhythmia Classifier

Authors

  • Muhammad Ilham Rizqyawan Indonesian Institute of Sciences, Indonesia Author
  • Artha Ivonita Simbolon Indonesian Institute of Sciences, Indonesia Author
  • Dwi Esti Kusumandari Indonesian Institute of Sciences, Indonesia Author

Keywords:

Android, Arrhythmia, ECG, SVM

Abstract

ECG is one of the most popular fields in biosignals research. One of the popular area in ECG research is automatic Arrhythmia classification. In this paper, we presented an effort to make an Arrhythmia classifier for Android. We use RRI based features and SVM as the classification method. Then we conduct an experiment with three different SVM configuration to see how much improvement can be made by using these configurations. By looking at kappa score as the metrics, the configuration 2 is greatly improve the classifier (169% increase). And by using hyper-parameter tuning we further optimize the classifier as can be seen on result of configuration 3 (10.5% increase).

Downloads

Published

2018-12-01

Issue

Vol. 10 No. 2 (2018): IIJ Volume 10, Number 2 (2018)

Section

Articles

How to Cite

Weighted SVM with RR Interval based Features for Android-based Arrhythmia Classifier. (2018). Internetworking Indonesia Journal, 10(2), 57-62. https://internetworkingindonesia.org/index.php/iij/article/view/130