Development of an Intelligent Agent that Detects Student's Negative Affect while Making a Computer Program

Programming is a fundamental course in IT education. Students who are new to programming are having difficulty in the course. Hence, the teachers' goal is to let students grasp the principles through lecture classes and let them apply these principles through laboratory exercises. Teachers guide students while making program by monitoring their activity. However, large laboratory classes are difficult to monitor every student's work. Thus, some students who need assistance, especially those who are confused, bored or timid to ask help are not all monitored. One way to address this is to develop an intelligent agent that detects student's negative affect and alerts the teacher once such affect is detected. By adopting the results of the study of reference [9], we were able to develop a system that implements an intelligent agent that monitors student's affect while making a computer program. The system was tested in a computer laboratory class with forty (40) client terminals. We observed that though the agent could hardly detects confusion state, it significantly detect and report boredom state and perfectly detect students who are consistently using other software applications that are not associated to the activity (off-task).