Organized by the International Working Group on Educational Data Mining.


Université du Québec à Montréal (UQAM), Canada

Machine Learning Department at the School of Computer Science, Carnegie Melon University

Full proceedings are available

Conference Schedule

All sessions held in SH-2420, Pavillon Sherbrooke, 200, rue Sherbrooke Ouest (on the UQAM campus)

Friday, June 20, 2008

9.00‑9.30Opening ceremony
9.30‑10.30Keynote address: Prof. Brian Junker, Assessment Modeling in Educational Data Mining
10.30‑10.45Coffee break
10.45‑12.15Session: EDM for Assessment
(30 mins each)
 Argument graph classification with Genetic Programming and C4.5
Collin Lynch, Kevin Ashley, Niels Pinkwart and Vincent Aleven

Mining Data from an Automated Grading and Testing System by Adding Rich Reporting Capabilities
Anthony Allevato, Matthew Thornton, Stephen Edwards and Manuel Perez-Quinones

 Adaptive Test Design with a Naive Bayes Framework
Michel Desmarais, Alejandro Villarreal and Michel Gagnon
13.35‑15.35Session: EDM for Improving Skill and Domain Models
(30 mins each)
 Labeling Student Behavior Faster and More Precisely with Text Replays
Ryan Baker and Adriana de Carvalho
 Acquiring Background Knowledge for Intelligent Tutoring Systems
Claudia Antunes
 Mining Student Behavior Models in Learning‑by‑Teaching Environments
Hogyeong Jeong and Gautam Biswas
 Integrating Knowledge Gained From Data Mining With Pedagogical Knowledge
Roland Hubscher and Sadhana Puntambekar
15.35‑15.50Coffee break
15.50‑17.20Session: Best Paper Nominees
(30 mins each)
 Data‑driven modelling of students' interactions in an ILE
Manolis Mavrikis
 A Response Time Model for Bottom‑Out Hints as Worked Examples
Benjamin Shih, Kenneth Koedinger and Richard Scheines
 Interestingness Measures for Association Rules in Educational Data
Agathe Merceron and Kalina Yacef



Saturday, June 21, 2008

9.00‑10.00Keynote address: Prof. David Pritchard, Assessing Learning
10.00‑10.15Coffee break
10.15‑11.00Young Researchers Track
(5 mins each)
 Skill Set Profile Clustering Based on Weighted Student Responses
Elizabeth Ayers, Rebecca Nugent and Nema Dean
 Developing a Log-based Motivation Measuring Tool
Arnon Hershkovitz and Rafi Nachmias
 Can we predict which groups of questions students will learn from?
Mingyu Feng, Neil Heffernan, Joseph Beck and Ken Koedinger
 Reinforcement Learning-based Feature Selection For Developing Pedagogically Effective Tutorial Dialogue Tactics
Min Chi, Pamela Jordan, Kurt VanLehn and Moses Hall
 Computational Infrastructures for School Improvement: A Way to Move Forward
Ben Shapiro, Hisham Petry and Louis Gomez
 Do Students Who See More Concepts in an ITS Learn More?
Moffat Mathews and Tanja Mitrovic
 Skill Mining Free-form Spoken Responses to Tutor Prompts
Xiaonan Zhang, Jack Mostow, Nell Duke, Christina Trotochaud, Joseph Valeri and Albert Corbett
 A Preliminary Analysis of the Logged Questions that Students Ask in Introductory Computer Science
Cecily Heiner
 Argument Mining Using Highly Structured Argument Repertoire
Safia Abbas and Hajime Sawamura
11.00‑12.00Poster session
13.15‑15.15Session: Improving Understanding of Student and Tutor Behaviors Through EDM
(30 mins each)
 Analytic Comparison of Three Methods to Evaluate Tutorial Behaviors
Jack Mostow and Xiaonan Zhang
 Using Item-type Performance Covariance to Improve the Skill Model of an Existing Tutor
Philip Pavlik, Hao Cen, Lili Wu and Ken Koedinger
 The Composition Effect: Conjuntive or Compensatory? An Analysis of
Multi-Skill Math Questions in ITS
Zachary Pardos, Joseph Beck, Neil Heffernan and Carolina Ruiz
 Improving Contextual Models of Guessing and Slipping with a Truncated Training Set
Ryan Baker, Albert Corbett and Vincent Aleven
15.15‑15.30Coffee break
15.30‑17.00Session: Tools to Support EDM
(30 mins each)
 Data Mining Algorithms to Classify Students
Cristobal Romero, Sebastián Ventura, Pedro G. Espejo and Cesar Hervas
 Can an Intelligent Tutoring System Predict Math Proficiency as Well as a Standardized Test?
Mingyu Feng, Joseph Beck, Neil Heffernan, and Ken Koedinger

 An open repository and analysis tools for fine-grained, longitudinal learner data
Ken Koedinger, Kyle Cunningham, Alida Skogsholm and Brett Leber
17.00‑17.30Closing Ceremony, Award