Predicting academic emotions based on brainwaves, mouse behaviour and personality profile

Authors:
Judith Azcarraga;Merlin Teodosia Suarez
Affiliations:
Center for Empathic Human-Computer Interactions, De La Salle Universtiy, Manila, Philippines;Center for Empathic Human-Computer Interactions, De La Salle Universtiy, Manila, Philippines
Venue:
PRICAI'12 Proceedings of the 12th Pacific Rim international conference on Trends in Artificial Intelligence
Year:
2012

Citing 6
Cited 1

The Aplusix-Editor: A New Kind of Software for the Learning of Algebra

ITS '02 Proceedings of the 6th International Conference on Intelligent Tutoring Systems
An Affective Model of Interplay between Emotions and Learning: Reengineering Educational Pedagogy-Building a Learning Companion

ICALT '01 Proceedings of the IEEE International Conference on Advanced Learning Technologies
Empirically building and evaluating a probabilistic model of user affect

User Modeling and User-Adapted Interaction
The WEKA data mining software: an update

ACM SIGKDD Explorations Newsletter
Emotion Sensors Go To School

Proceedings of the 2009 conference on Artificial Intelligence in Education: Building Learning Systems that Care: From Knowledge Representation to Affective Modelling
Use of Personality Profile in Predicting Academic Emotion Based on Brainwaves Signals and Mouse Behavior

KSE '11 Proceedings of the 2011 Third International Conference on Knowledge and Systems Engineering

Recognition of understanding level and language skill using measurements of reading behavior

Proceedings of the 19th international conference on Intelligent User Interfaces

Quantified Score

Hi-index	0.00

Visualization

Abstract

This research presents how the level of academic emotions such as confidence, excitement, frustration and interest, may be predicted based on brainwaves and mouse behaviour, while taking into account the student's personality. Twenty five (25) college students of different personalities were asked to use the Aplusix® algebra learning software while an EEG sensor was attached to their head to capture their brainwaves. Brainwaves were carefully synchronized with the mouse behaviour and the assigned student activity. The collected brainwaves were then filtered, pre-processed and transformed to different frequency bands (alpha, beta, gamma). A number of classifiers were then built using different combinations of frequencies and mouse information which were used to predict the intensity level (low, average, high) of each emotion.