Automatic generation of concept maps from students’ answers (1)
Concept maps can be a basis for a recommendation system for questions. One possibility is to generate them directly from students' answers. The underlying principle is the assumption that all concepts can be categorized into two groups: 1. concepts that students have mastered, and 2. concepts that students have not mastered. The criterion “mastered” is optional. Applying this concept to our data has shown that it is not applicable (students’ patterns are not stable enough to be learnable ).
Example:
Parse our course questions and student responses to them (how they answered questions ). Construct concept map as advised by S.-M. Chen and P.-J. Sue.
Reference:
[1] S.-M. Chen and S.-M. Bai, “Using data mining techniques to automatically construct concept maps for adaptive learning systems,” Expert Systems with Applications, vol. 37(6 ), p. 4496–4503, Jun 2010.
[2] S.-M. Chen and P.-J. Sue, “Constructing concept maps for adaptive learning systems based on data mining techniques,” Expert Systems with Applications, vol. 40(7 ), p. 2746–2755, Jun 2013.
[3] F. Krieglstein, S. Schneider, M. Beege, and G. D. Rey, “How the design and complexity of concept maps influence cognitive learning processes,” Educational Technology Research and Development, vol. 70(1 ), pp. 99–118, Feb 2022.
[2] S.-M. Chen and P.-J. Sue, “Constructing concept maps for adaptive learning systems based on data mining techniques,” Expert Systems with Applications, vol. 40(7 ), p. 2746–2755, Jun 2013.
[3] F. Krieglstein, S. Schneider, M. Beege, and G. D. Rey, “How the design and complexity of concept maps influence cognitive learning processes,” Educational Technology Research and Development, vol. 70(1 ), pp. 99–118, Feb 2022.
Author of the tip:
Andrej Košir
University of Ljubljana