Student Fatigue and Attention Mental State Recognition using EEG Brain Signal Classifier

Recently, brain-computer interface (BCI) field has allowed human mental state to be recognized and classified by using non-invasive brain signal; Electroencephalogram (EEG). In capturing brain signal, invasive approach such as ECoG provides higher temporal and a lower vulnerability of artifacts compared to non-invasive. However, ECoG requires a surgical perforation to implant an electrode grid on the brain. Thus, this approach is not practical to be implemented in classroom. Consequently, the convenient and practical way of capturing electrical brain signal in classroom is through EEG by using non-invasive portable and wearable device at cheaper cost like NeuroSky. However, the down side of capturing EEG signal using wearable device is it offers poor quality signal as it is rigorously affected by noise. Thus, we proposed an automated system uses advanced filtering algorithm and classification techniques to identify students’ mental state as fatigue or attentive by using EEG signal. The objectives of this project are to propose a filtering algorithm to reduce noise in EEG signal, to propose an enhanced classification algorithm for fatigue and attention and to develop a prototype for automatic classification. The methodology consists of 6 main steps; 1) EEG signal acquisition: Raw EEG signal is collected, 2) Pre-processing EEG signal: The poor quality of EEG signals is filtered by applying linear and adaptive based filtering algorithm, 3) Features extraction: Clean signal feature for classification stage, 4) Classification: EEG signals is classified as fatigue and attention mental state by applying LR and MLP classifier as, 5) Experimental: Classroom experimental session for validating the proposed work, 6) Prototype development: A working product is produced for classroom activities. The expected output is designing

an enhanced filtering algorithm and enhanced EEG classifier for classifying students’ mental state. Finally, improving student understanding of subject as well as improving the national education policy and standard. 

In this work, we proposed an automated computer system framework that uses advanced filtering algorithm and classification techniques to identify student’ mental state which are fatigue or attentive using EEG signal.

1. To propose a linear and adaptive based filtering algorithm to reduce noise in EEG brain signal.
2. To propose an enhanced classification algorithm for fatigue and attention EEG signal.
3. To examine the optimum metric values for fatigue and attention experiment and valuation.
4. To develop a prototype for automatic classification of EEG signal to detect mental state in student dataset acquired by NeuroSky. 

 Publication

Sumari, P., Abdullah, Jafri, M., Idris, Z., (2017), “We must invest in applied knowledge of computation al neurosciences and informatics as an important future in Malaysia: The Malaysian brain mapping project”, in Malaysian Journal of Medical Science”

  • Hits: 1483

School of Computer Sciences, Universiti Sains Malaysia, 11800 USM Penang, Malaysia
Tel: +604-653 3647 / 2158 / 2155  |  Fax: +604-653 3684  | Email: This email address is being protected from spambots. You need JavaScript enabled to view it.  |  icon admin