Comparison of feature selection methods in ECG signal classification

Authors:
Zhen-Xing Zhang;Sang-Hong Lee;Joon S. Lim
Affiliations:
Kyungwon University, Sungnam, Korea;Kyungwon University, Sungnam, Korea;Kyungwon University, Sungnam, Korea
Venue:
Proceedings of the 4th International Conference on Uniquitous Information Management and Communication
Year:
2010

Citing 6
Cited 2

Performance comparison of several feature selection methods based on node pruning in handwritten character recognition

ICDAR '97 Proceedings of the 4th International Conference on Document Analysis and Recognition
A novel feature selection method to improve classification of gene expression data

APBC '04 Proceedings of the second conference on Asia-Pacific bioinformatics - Volume 29
Principal component analysis in ECG signal processing

EURASIP Journal on Applied Signal Processing
Feature selection using principal feature analysis

Proceedings of the 15th international conference on Multimedia
Extracting Fuzzy Rules for Detecting Ventricular Arrhythmias Based on NEWFM

PAKDD '09 Proceedings of the 13th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining
Finding features for real-time premature ventricular contraction detection using a fuzzy neural network system

IEEE Transactions on Neural Networks

A fatigue detection algorithm by heart rate variability based on a neuro-fuzzy network

Proceedings of the 5th International Conference on Ubiquitous Information Management and Communication
Application of machine learning techniques to analyse the effects of physical exercise in ventricular fibrillation

Computers in Biology and Medicine

Quantified Score

Hi-index	0.00

Visualization

Abstract

In past years, the Principal Component Analysis (PCA) has been applied to select features for classification applications. This paper presents a performance comparison between PCA and Non-overlap Area Distribution Measurement (NADM), which is based on a neural fuzzy network. This paper performs an experiment on Normal Sinus Rhythm (NSR) and Ventricular Tachycardia/Fibrillation (VT/VF) classification with the two feature selection methods. The performance result is 89.34% while the number of initial features is projected from six to four by the PCA method. The performance result is 91.02% while the number of initial features is decreased from six to two by NADM. The results clearly show that NADM outperforms PCA by 1.68% with fewer features.