Decision Tree for Tracking Learner’s Emotional State Predicted from His Electrical Brain Activity
This paper proposes the use of machine learning techniques to build an efficient learner’s emotional transition diagram transition. For information Extraction tasks, we led an experimentation in which we exposed a group of 17 learners to a series of pictures from the International Affective Picture System (IAPS). Decision tree classifier has demonstrated the best ability to learn model structure from data collected. Among the emotions involved in learning and according to the picture from IAPS and the current emotional state, we drew up the transition diagram. Our model aims to improve the task of predicting the emotional state in an Intelligent Tutoring System and achieve a prediction accuracy of 63.11%. These results suggest that the implementation of the decision tree algorithm in the intelligent tutoring system we are developing improves the ability for an ITS to track the learners emotional states.
Unable to display preview. Download preview PDF.
- 1.D’Mello, S.K., Craig, S.D., Gholson, B., Franklin, S., Picard, R.W., Graesser, A.C.: Integrating Affect Sensors in an Intelligent Tutoring System. In: Affective Interactions: The Computer in the Affective Loop Workshop at 2005 International conference on Intelligent User Interfaces, pp. 7–13. AMC Press, New York (2005)Google Scholar
- 2.Heraz, A., Frasson, C.: Predicting the three major dimensions of the learner’s emotions from brainwaves. In: 4th International Conference on Computational Intelligence and Cognitive Informatics: CICI 2007, Venise, Italy (2007)Google Scholar
- 3.Heraz, A., Razaki, R., Frasson, C.: Using machine learning to predict learner emotional state from brainwaves. In: 7th IEEE conference on Advanced Learning Technologies: ICALT 2007, Niigata, Japan (2007)Google Scholar