Issue No. 04 - July/August (2007 vol. 22)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MIS.2007.73
Vivien Robinet , TIMC-IMAG Laboratory
Gilles Bisson , TIMC-IMAG Laboratory
Mirta B. Gordon , TIMC-IMAG Laboratory
Benoît Lemaire , TIMC-IMAG Laboratory
This article applies machine-learning techniques to student modeling, presenting a method for discovering high-level student behaviors from a very large set of low-level traces corresponding to problem-solving actions in a learning environment. The system encodes basic actions into sets of domain-dependent attribute-value patterns. Then, a domain-independent hierarchical clustering identifies high-level abilities, yielding natural-language diagnoses for teachers. The method can be applied to individual students or to entire groups, such as a class. The system was applied to the actions of thousands of students in the domain of algebraic transformations. This article is part of a special issue on intelligent educational systems.
computer-assisted instruction, machine learning, education, data mining
M. B. Gordon, B. Lemaire, G. Bisson and V. Robinet, "Inducing High-Level Behaviors from Problem-Solving Traces Using Machine-Learning Tools," in IEEE Intelligent Systems, vol. 22, no. , pp. 22-30, 2007.