Using data mining on student behavior and cognitive style data for improving e-learning systems: a case study

International Journal of Computational Intelligence Systems, Vol. 5, No. 3 (June, 2012), 597-610 Using data mining on student behavior and cognitive style data for improving e-learning systems: a case study Milos Jovanovic, Milan Vukicevic, Milos Milovanovic, Miroslav Minovic Faculty of Organizational Sciences, University of Belgrade, Jove Ilica 154 Belgrade, Serbia E-mail: {milos.jovanovic, milan.vukicevic, milos.milovanovic, miroslav.minovic}@fon.bg.ac.rs www.bg.ac.rs Received 6 December 2011 Accepted 15 May 2012 Abstract In this research we applied classification models for prediction of students’ performance, and cluster models for grouping students based on their cognitive styles in e-learning environment. Classification models described in this paper should help: teachers, students and business people, for early engaging with students who are likely to become excellent on a selected topic. Clustering students based on cognitive styles and their overall performance should enable better adaption of the learning materials with respect to their learning styles. The approach is tested using well-established data mining algorithms, and evaluated by several evaluation measures. Model building process included data preprocessing, parameter optimization and attribute selection steps, which enhanced the overall performance. Additionally we propose a Moodle module that allows automatic extraction of data needed for educational data mining analysis and deploys models developed in this study. Keywords: educational data mining, prediction, students, performance, classification, clustering, Moodle. 1. Introduction Moodle is an open source Learning Management System (LMS) that is mostly regarded as Course Management System by the open community. It is dominantly used in higher education and it has proven as a successful tool in that setting.1,2 For that reason our faculty built a distance learning system (DLS) based on Moodle LMS. The system was built and developed as an in-house solution at University of Belgrade for the students of Information technology. One of the main requirements was to completely support distance learning process in all its aspects. The system enables dealing with advanced courses, which use multimedia lessons, advanced workshops and face to face communication through video conferencing. Web-based learning management systems are extensively used nowadays and produce vast amounts of data that are potentially useful for improving educational process.2,4,5 The new emerging field, called Educational Data Mining (EDM), concerns with developing methods that discover knowledge from data originating from educational (traditional or distance learning) environments.6 Increasing research interests in using data mining in education is recorded in the last decade7,8,9,10,11 with focus on different aspects of educational process (e.g. students, teachers, teaching materials, organization of classes etc.). Benefits from extracting knowledge from e-learning data are expected under assumption that the trails of user actions can be used to identify specific information on users. We hope that the user behavior captured in log files and recorded in data structures can be used to create models that predict user behavior, or describe their peculiarities. There are several groups of people who can leverage this knowledge, and are potential stakeholders: Students, Teachers, e-learning system administrators, University management. These stakeholders could use this knowledge for different goals9: Published by Atlantis Press Copyright: the authors 597 Milos Jovanovic, Milan Vukicevic , Milos Milovanovic, Miroslav Minovic 1. Applications dealing with the assessment of students’ learning performance. 2. Applications that provide course adaptation and learning recommendations based on the students’ learning behavior. 3. Approaches dealing with the evaluation of learning material and educational web based courses. 4. Applications that involve feedback to both teachers and students of e-learning courses, based on the students’ learning behavior. 5. Developments for the detection of atypical students’ learning behavior. These goals are achieved with help of data mining techniques such as k-nearest neighbor, naive Bayes, decision trees, artificial neural networks, support vector machines, K-means, hierarchical clustering etc.12 Still, learning management systems are not primarily designed with data analysis and mining in mind, because usage data is not stored in a systematic way. Its thorough analysis requires long and tedious preprocessing.13 Furthermore, LMS systems usually produce statistic reports. These reports however do not assist instructors in drawing out useful conclusions either for the course potential or student abilities and are useful only for platform administrative purposes.2 This research shows how one can leverage the available data on student behavior, in order to predict success of students, as well as profile students into groups which may help improve existing learning material and collaborative learning. The study involves data from students attending online (distance learning) university courses as suggested by Romero et al.,6 and extends available data with students cognitive styles. Additionally we propose Moodle module that allows automatic extraction of data needed for EDM analysis and deploys models evolved in this study. The paper is structured as follows: Section 2 introduces related work on using e-learning data and applying data mining models. Architectural design of the decisionsupport system is given in Section 3, with experimental results in using data mining models presented in Section 4. Potential ways of using knowledge gained by data mining models is described in Section 5, and Section 6 discusses open issues and related problems for these types of applications. 2. Background Romero and Ventura gave a systematic survey about EDM from 1995 to 2005.10 Because of increasing popularity and number of researches in this area, the same authors gave an extensive overview about the state of the art in this area until 2011 with over 300 references.12 In this paper we will focus on researches that are closest to our work. Study by Wang and Liao was performed in order to investigate how Data Mining techniques can be successfully used for adaptive learning.14 In academic institutions, Moodle platform is often utilized as a significant part of e-learning systems. Romero et al. described how different data mining techniques can be used in that setting to improve the course and the students’ learning.6 Applications or tasks that have been resolved through data mining techniques are classified by Romero and Ventura in twelve categories: Analysis and visualization of data, Providing feedback for supporting instructors, Recommendations for students, Predict students’ performance, Student modeling, Detecting undesirable student behaviors, Grouping students S (...truncated)