An Implementation of FBCSP Replacing Feature Selection Method According to the Classifier

DOI

10.2991/wartia-16.2016.349How to use a DOI?

Keywords

MI signal, BCI, FBCSP, Relief, FLD

Abstract

Common spatial pattern (CSP) is a popular method for 2-class motor imagery electroencephalogram (EEG) classification. Filter bank common spatial pattern (FBCSP) is an improved version of CSP by introducing multiple frequency filters and a stage of feature selection. In FBCSP, methods for the feature selection stage can be various, and there is no criteria to guide the selection. To make the selection of a specific feature selection method reasonable, this paper tries to obey a guideline that the feature selection method should match with the classification method. Based on the proposed criteria above, this paper employs fisher linear discriminant (FLD) for classification and a corresponding Relief method for features selection. Introduction

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Kang Wang
AU  - Xueqian Wang
AU  - Gang Li
AU  - Bin Liang
PY  - 2016/05
DA  - 2016/05
TI  - An Implementation of FBCSP Replacing Feature Selection Method According to the Classifier
BT  - Proceedings of the 2016 2nd Workshop on Advanced Research and Technology in Industry Applications
PB  - Atlantis Press
SP  - 1761
EP  - 1765
SN  - 2352-5401
UR  - https://doi.org/10.2991/wartia-16.2016.349
DO  - 10.2991/wartia-16.2016.349
ID  - Wang2016/05
ER  -