Association Analysis on Open-Ended Concept Maps using Data Mining

Abstract

An open-ended concept map is an appealing approach that allows learners to add concepts and links freely, representing their understanding. This technique reveals differences among students and accurately captures their knowledge structure. However, manually evaluating an open-ended map is challenging and time-consuming, particularly in a large classroom, and involves many propositions. Several works tried to use data mining techniques to generate concept maps from the source text rather than analyzing the knowledge structure. The previous study employed association analysis on provided concept maps. This study focused on association analysis of open-ended concept mapping in revealing hidden students’ understanding. A total of 20 open-ended concept maps were involved in this study. The findings suggested that data mining techniques could quickly disclose necessary information on openended concept maps. The study identified an exciting topic for further research in educational data mining, particularly concerning knowledge structure.