Mining learner profile utilizing association rule for web-based learning diagnosis

With the rapid growth of computer and Internet technologies, c-learning has become a major trend in the computer assisted teaching and learning fields. Most past researches for web-based learning focused on the issues of adaptive presentation, adaptive navigation support, curriculum sequencing, and intelligent analysis of student's solutions. These systems commonly neglect to consider whether learner can understand the learning courseware and generate misconception or not. To neglect learner's learning misconception will lead to obviously reducing learning performance, thus generating learning difficult. In order to discover common learning misconceptions of learners, this study employs the association rule to mine the learner profile for diagnosing learners' common learning misconceptions during learning processes. In this paper, the association rules that occurring misconception A implies occurring misconception B can be discovered utilizing the proposed association rule learning diagnosis approach. Meanwhile, this study applies the discovered association rules of the common learning misconceptions to tune courseware structure through modifying the difficulty parameters of courseware in the courseware database so that learning pathway is appropriately tuned. Besides, this paper also presents a remedy learning approach based on the discovered common learning misconceptions to promote learning performance. Experiment results indicate that applying the proposed learning diagnosis approach can correctly discover learners' common learning misconceptions according to learner profile and help learners to learn more effectively. (c) 2006 Elsevier Ltd. All rights reserved.