Classification of Mental Task From EEG Signals Using Immune Feature Weighted Support Vector Machines

被引:56
作者
Guo, Lei [1 ]
Wu, Youxi [1 ]
Zhao, Lei [2 ,3 ,4 ]
Cao, Ting [1 ]
Yan, Weili [1 ]
Shen, Xueqin [1 ]
机构
[1] Hebei Univ Technol, Prov Minist Joint Key Lab Elect Field & Elect App, Tianjin 300130, Peoples R China
[2] Harvard Univ, Sch Med, Dept Radiol, Boston, MA 02115 USA
[3] Brigham & Womens Hosp, Boston, MA 02115 USA
[4] XinAoMDT Technol Co Ltd, Langfang 065001, Hebei, Peoples R China
关键词
Feature weight; immune algorithm; mental task; support vector machine; ALGORITHM;
D O I
10.1109/TMAG.2010.2072775
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The classification of mental tasks is one of key issues of EEG-based brain computer interface (BCI). Differentiating classes of mental tasks from EEG signals is challenging because EEG signals are nonstationary and nonlinear. Owing to its powerful capacity in solving nonlinearity problems, support vector machine (SVM) method has been widely used as a classification tool. Traditional SVMs, however, assume that each feature of a sample contributes equally to classification accuracy, which is not necessarily true in real applications. In addition, the parameters of SVM and the kernel function also affect classification accuracy. In this study, immune feature weighted SVM (IFWSVM) method was proposed. Immune algorithm (IA) was then introduced in searching for the optimal feature weights and the parameters simultaneously. IFWSVM was used to multiclassify five different mental tasks. Theoretical analysis and experimental results showed that IFWSVM has better performance than traditional SVM.
引用
收藏
页码:866 / 869
页数:4
相关论文
共 13 条
[1]  
[Anonymous], P 7 INT C ART IMM SY
[2]  
Balakrishnan D, 2005, NORTHEAST BIOENGIN C, P118
[3]   A tutorial on Support Vector Machines for pattern recognition [J].
Burges, CJC .
DATA MINING AND KNOWLEDGE DISCOVERY, 1998, 2 (02) :121-167
[4]   An immune algorithm for protein structure prediction on lattice models [J].
Cutello, Vincenzo ;
Nicosia, Giuseppe ;
Pavone, Mario ;
Timmis, Jonathan .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2007, 11 (01) :101-117
[5]   Support vector EEG classification in the Fourier and time-frequency correlation domains [J].
Garcia, GN ;
Ebrahimi, T ;
Vesin, JM .
1ST INTERNATIONAL IEEE EMBS CONFERENCE ON NEURAL ENGINEERING 2003, CONFERENCE PROCEEDINGS, 2003, :591-594
[6]   BCI competition 2003 - Data set IIb: Support vector machines for the P300 speller paradigm [J].
Kaper, M ;
Meinicke, P ;
Grossekathoefer, U ;
Lingner, T ;
Ritter, H .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2004, 51 (06) :1073-1076
[7]   A NEW MODE OF COMMUNICATION BETWEEN MAN AND HIS SURROUNDINGS [J].
KEIRN, ZA ;
AUNON, JI .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 1990, 37 (12) :1209-1214
[8]   EEG-based synchronized brain-computer interfaces: A model for optimizing the number of mental tasks [J].
Kronegg, Julien ;
Chanel, Guillaume ;
Voloshynovskiy, Sviatoslav ;
Pun, Thierry .
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, 2007, 15 (01) :50-58
[9]  
Li ZW, 2007, ICEMI 2007: PROCEEDINGS OF 2007 8TH INTERNATIONAL CONFERENCE ON ELECTRONIC MEASUREMENT & INSTRUMENTS, VOL III, P906
[10]   APPROXIMATE ENTROPY AS A MEASURE OF SYSTEM-COMPLEXITY [J].
PINCUS, SM .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 1991, 88 (06) :2297-2301