Research Assistant Intern 0923
January 08, 2009
Project Name
Modeling Learner Affective States in Intelligent Tutoring Systems
Project Description
In recent years, the ITS community has increasing recognized the importance of addressing learner’s affective needs. During the recent Life-long Learning Companion workshop held at ICT, understanding and addressing learner and tutor affect were considered by the participants to be essential to building learning companions that can establish and maintain long-term relations with the learner. This project aims to gain better understanding of learner affective states in intelligent tutoring systems (possibly educational games). In particular, the project would investigate the potential of using appraisal theories of emotion, e.g. EMA, to model learner affective states in intelligent tutoring systems. This investigation would further benefit the existing tutoring systems at ICT.
Job Description
• Implement data collection module in an existing tutoring system.
• Conduct a user study to collect data of user interactions in the tutoring system and user’s self-assessment of goals and emotions while using the tutoring system.
• Using machine learning to create a model of user’s goals and different appraisal dimensions.
• Using the goals and appraisal dimensions to build a computational model of learner emotions.
Skill set:
• Good C/C++ programming skills
• Familiar with machine learning
• Familiar with plan recognition
• Familiar with HMM.