A "Frequently Asked Questions" Management System that Supports Voting, Built for Student Evaluation and Optimization Purposes Huu Le Van, Andrea Trentini, Universita' di Milano Bicocca, Italy {levan,trentini}@disco. u nimib, it We have built a system for managing F A Q lists in a special way, to be used in an educational environment. Our system supports an explicit voting mechanism to let users assess the relative importance o f any posted question. Our main goals are: 1. 2. to optimize the student-teacher interaction; to evaluate the effectiveness o f learning. their knowledge progress on their guesses. W e can calculate a student performance in learning by giving values to his "actions" (answers/votes). We can give values to actions by studying the behavior o f the whole group (e.g. a very difficult question will have more value if correctly answered and tess value i f wrongly answered) W e are also studying the possibility to use some semantic distance algorithm to collapse questions/answers too near to be treated as different. This "collapsing feature" could be used as another evaluation technique: students posting questions/answers that do not "augment" the database must be discouraged. The implementation is very simple: a "weighted" stack o f questions, in which every item gains weight by being voted. A web server interface (built around Java Servlets) serves as access point from users. Users must register themselves, since every posting is marked with the ID o f the poster. As in a standard F A Q management system, in our F A Q sharing system students can contribute to a F A Q collection on the subject taught. They can post questions and answers, but then they can also vote both types o f items (questions for interest, answers for correctness). This way we can optimize the time spent in answering to their doubts, because we only answer questions THEY consider important. The interesting part is also the evaluation part. Since we also let them try to answer questions and vote for answers, we can study The Logic Tutor David Abraham, Liz Crawford, Leanna Lesta, Agathe Merceron and Kalina Yacef, Basser Department of Computer Science, University of Sydney, Australia {dabraham,Uzc,llesta}@ug.cs.usyd.edu.au {agathe,kalina}@cs.usyd.edu.au This paper presents a Logic Tutor, a tool to support computer science students in their learning of logic and more specifically in their learning o f formal proofs. The current tool is equipped with a deduction system for propositional logic. However its modular conception makes it easy to change to another logic. Preliminary evaluation shows that this tool has a high educational value, thanks, among others features, to its simple, attractive interface, and its highly specific error messages. Because o f this promising evaluation, the Logic Tutor will be integrated in our Logic teaching course in 2001.
/lp/association-for-computing-machinery/a-frequently-asked-questions-management-system-that-supports-voting-LTlKX7aEOZ