IIUM Repository

Evaluation of feature extraction and classification techniques for EEG-based subject identification

Handayani, Dini Oktarina Dwi and Abdul Rahman, Abdul Wahab and Yaacob, Hamwira Sakti (2016) Evaluation of feature extraction and classification techniques for EEG-based subject identification. Jurnal Teknologi, 78 (9). pp. 41-48. E-ISSN 2180–3722

[img] PDF - Published Version
Restricted to Repository staff only

Download (852kB) | Request a copy
[img] PDF - Published Version
Restricted to Repository staff only

Download (138kB) | Request a copy
[img] PDF - Published Version
Restricted to Repository staff only

Download (171kB) | Request a copy

Abstract

The ability to identify a subject is indispensable in affective computing research due to its wide range of applications. User profiling was created based on the strength of emotional patterns of the subject, which can be used for subject identification. Such system is made based on the emotional states of happiness and sadness, indicated by the electroencephalogram (EEG) data. In this paper, we examine several techniques used for subject profiling or identification purposed. Those techniques include feature extraction and classification techniques. In the experimental study, we compare three techniques for feature extraction namely, Power Spectral Density (PSD), Kernel Density Estimation (KDE), and Mel Frequency Cepstral Coefficients (MFCC). As for classification we compare three classification techniques, they are; Multilayer Perceptron (MLP), Naive Bayesian (NB), and Support Vector Machine (SVM). The best result achieved was 59.66%, using the MFCC and MLP-based techniques using 5-fold cross validation. The experiment results indicated that these profiles could be more accurate in identifying subject compared to NB and SVM. The comparisons demonstrated that profile-based methods for subject identification provide a viable and simple alternative to this problem.

Item Type: Article (Journal)
Additional Information: 4870/56434
Uncontrolled Keywords: Subject Identification, Power Spectral Density, Kernel Density Estimation, Mel Frequency Cepstral Coefficients, Multilayer Perceptron, Naive Bayesian, and Support Vector Machine
Subjects: T Technology > T Technology (General)
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button): Kulliyyah of Information and Communication Technology > Department of Computer Science
Kulliyyah of Information and Communication Technology > Department of Computer Science
Depositing User: Dr Hamwira Yaacob
Date Deposited: 12 Apr 2017 09:58
Last Modified: 18 Apr 2017 09:28
URI: http://irep.iium.edu.my/id/eprint/56434

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year