skill model, association rules, cliques

In order to evaluate students? progress in education, a didactical testing takes place, on which pedagogical and psychological experts define key for calculating the level of individual student skills. Such model is based on assigning testing questions to defined skills. The aim of this paper is to explore possibilities of using data mining feature extraction methods to extract skill models from didactical tests. An experiment is performed that retrieves skill model based on cliques found in association rules.