About This Project

Ask the Scientists

What is the context of this research?

This research proposes to expand on the existing CS education intelligent Tutoring System, Chi-Qat-Tutor. Chi-Qat, formerly known as iList, aims at facilitating learning of basic CS data structures (e.g., linked lists, trees, stacks) and algorithmic strategies. Both widely accepted pedagogical techniques and those compiled through analysis of an accompanying tutoring corpus guide the ITS.

Chi-Qat’s primary modules involves the subject of linked list, recursion, and binary search trees. The desktop-based, Java implemented system presents students with a problem statement along with a simulated environment in which they can visualize and programmatically alter data. The most recent version of the system has produced learning gains that surpass that of human tutors.

What is the significance of this project?

A computer science intelligent tutoring system, combined with the assets of a mobile device, has the potential to facilitate richer and more accessible CS learning. However, there has been little exploration of a mobile ITS intended to better CS education.

The design of a CS education mobile ITS undoubtedly requires reconsideration of the graphical interface. Further, mobile devices offer an avenue for exploration of tutor feedback that differs from a desktop system. Namely, the usage of the user location and other aspects of user context may benefit feedback synthesis. Finally, the addition of collaborative problem solving will reinforce learning and prepare students for real-world team-based problem solving.

This research takes a needed & promising step toward furthering CS education.

What are the goals of the project?

1. Introduce Collaboration Functionality
   • Connected devices are pathways to other learners with expertise
   • Peer-to-peer question and answer capability will be added to the current tutoring system
   • Learners may post their uncertainties regarding code
   • Peers may in turn respond with their expert advice
   • This collaboration will also enable reinforcement learning through filtering of peer questions such that those who have recently encountered the same issue are encouraged to provide expertise.
   • The optimal selection of peers to form these collaborative networks will also be explored
2. Complete Mobile System Development
   
   • Currently ChiQat-Tutor is desktop based
   • Work is underway to develop the mobile version of the system for Android Tablets
   • Feedback and problems are adapted directly from the desktop version