Exploring machine learning approach for attention detection using public EEG datasets

Handayani, Dini Oktarina Dwi and Yaacob, Hamwira Sakti and Othman, Marini and Basri, Atikah Balqis and Sulistiani, Heni and Darwis, Dedi and Isnin, Megat Arif Ilham and Maslan, Mas Azlan Hafiz (2026) Exploring machine learning approach for attention detection using public EEG datasets. In: 10th International Conference on Information and Communication Technology for the Muslim World, ICT4M 2025, 26-27 November 2025, Kuala Lumpur.

	PDF - Published Version Restricted to Registered users only Download (1MB) | Request a copy
Preview	PDF - Supplemental Material Download (157kB) | Preview

Abstract

Brain Computer Interface (BCI) technology provides new opportunities for objectively assessing cognitive states such as attention, which plays a critical role in effective learning. In today's digital and hybrid classrooms, students often encounter distractions that hinder their engagement and performance. Meanwhile, conventional assessment methods rely heavily on subjective observations or self-reports, which limit realtime accuracy. This preliminary study examines the feasibility of predicting attention levels using electroencephalography (EEG) signals and machine learning techniques. A publicly available Kaggle EEG dataset, recorded from four electrodes (AF7, AF8, TP9, TP10), was preprocessed through feature extraction, normalisation, and partitioning into training and testing sets. An XGBoost regression model was trained to predict continuous concentration scores from multichannel EEG features. The model achieved a high performance with a Mean Squared Error (MSE) of 0.0031 and a coefficient of determination (R²) of 0.9764. These results demonstrate the potential of regression-based EEG analysis for real-time, continuous attention monitoring, offering finer granularity compared to traditional categorical approaches. The findings serve as an initial step toward developing TARKEEZ, a real-time EEG-based brainwave monitoring system designed to enhance student engagement and learning efficiency.

Item Type:	Proceeding Paper (Other)
Uncontrolled Keywords:	EEG, BCI, Attention Monitoring, XGBoost Regression, Concentration Prediction, Educational Technology
Subjects:	Q Science > QA Mathematics > QA76 Computer software
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Information and Communication Technology Kulliyyah of Information and Communication Technology Kulliyyah of Information and Communication Technology > Department of Computer Science Kulliyyah of Information and Communication Technology > Department of Computer Science
Depositing User:	Dr Dini Handayani
Date Deposited:	22 Apr 2026 09:25
Last Modified:	22 Apr 2026 09:25
Queue Number:	2026-04-Q2832
URI:	http://irep.iium.edu.my/id/eprint/128328

Actions (login required)

View Item