1479 - Reaction Explorer: Organic chemistry tutorial system with inherent predictive power and collaborative learning support
Jonathan H. Chen PhD, Pierre Baldi PhD Institute for Genomics and Bioinformatics, School of Information and Computer Sciences, University of California, Irvine, Irvine, CA, United States; School of Medicine, University of California, Irvine, Irvine, CA, United States
Reaction Explorer is an interactive tutorial system for organic chemistry that can generate a virtually limitless number of multi-step synthesis design and reaction mechanism problems with support for inquiry-based learning. This electronic tutor is powered by an underlying reaction expert system, comprising over 80 reagent models and 1,500 manually-curated reaction pattern rules, giving it inherent predictive power spanning the undergraduate organic chemistry curriculum.
While the system can generate a virtually limitless number of practice problems for students, research and development is ongoing to optimize the problem selection for students beyond "random" generation. This includes work to develop a knowledge model for organic chemistry to dynamically customize the content and difficulty of generated problems to match the real-time perceived capability of individual students. To further enhance the selection of generated problems, a crowd-sourcing framework is established to allow students to comment on and rate problems as they progress to help identify more interesting and useful generated problems, and even to allow users to submit their own custom-authored problems to share with the community for use and feedback.
Pedagogical experiments in undergraduate classes indicate that the system can improve average student examination performance by ~10%. The system is accessible via http://cdb.ics.uci.edu
Thursday, March 31, 2011 01:45 PM
Advances in Teaching Organic Chemistry (01:00 PM - 04:10 PM)
Location: Disney's Grand Californian Hotel
Room: Trillium B