This workshop will include a discussion of the 2010 KDD Cup competition, and the winning teams will present their work.
This year's competition asked participants to predict student performance on mathematical problems from logs of student interaction with Intelligent Tutoring Systems. The competition addressed questions of both scientific and practical importance. Improved models could be saving millions of hours of students' time (and effort) in learning algebra. These models should both increase achievement levels and reduce time needed to learn.
For more information, see the KDD Cup 2010 website.
Time | Duration | |
Intro | 9am | 5 min |
Invited Talk Cognitive Tutor: Modeling to improve mathematics education Steve Ritter, Co-Founder & Chief Scientist Tristan Nixon, Research Programmer Carnegie Learning, Inc. |
9:05am | 30 min |
Organizers' Talk John Stamper and Alexandru Niculescu-Mizil |
9:35am | 20 min |
Winning Talk: 1st Place Overall Feature engineering and classifier ensembling for KDD Cup 2010 National Taiwan University team (workshop paper) |
9:55am | 20 min |
Break | 10:15am | 20 min |
Winning Talk: 3rd Place Overall Collaborative Filtering Applied to Educational Data Mining Andreas Töscher, BigChaos @ KDD team (workshop paper) |
10:35am | 20 min |
Winning Talk: 2nd Place Student Track Using HMMs and bagged decision trees to leverage rich features of user and skill Zach A. Pardos (workshop paper) |
10:55am | 20 min |
Winning Talk: 3rd Place Student Track Split-Score-Predicate SCUT Data Mining team (workshop paper) |
11:15am | 20 min |
EDM Future Discussion Ken Koedinger |
11:35am | 20 min |
Closing | 11:55am | 5 min |
Poonam Gandhi & Varun Aggarwal Ensemble Hybrid Logit Model |
|
Tri Kurniawan Wijaya & Philips Kokoh Prasetyo Knowledge Tracing with Stochastic Method |
|
Yasser Tabandeh & Ashkan Sami Classification of Tutor System Logs with High Categorical Features |
|
Rafael Perez Mendoza, Neil Rubens & Toshio Okamoto Hierarchical Aggregation Prediction Method |
|
Kun Liu & Yan Xing A Lightweight Solution to the Educational Data Mining Challenge |