Optimizing response prediction by an interactive tutoring system

The aim of the project is to increase the intelligence of an automated, online tutoring system, using state-of-the-art techniques and concepts from cognitive science and computer science. This will be accomplished by integrating more data about each learner, and by implementing a more advanced model able to predict learner behaviour. The improved system will better be able to gauge and represent the knowledge levels of individual learners, based on their answers to questions posed by the system. The system will also be able to generate an appropriate study plan for each individual learner, consisting of a sequence of questions tailored to the learners knowledge level. The improved system should remain scalable, and able to serve many learners simultaneously in a real-time fashion.

Faculty Supervisor:

Sheila McIlraith

Student:

Vishal Raheja

Partner:

NeuRecall Inc.

Discipline:

Computer science

Sector:

Information and communications technologies

University:

University of Toronto