Document Type : Original Research Paper
Authors
Computer and IT Group, Department of Electrical and Computer Engineering, Isfahan University of Technology, Isfahan, Iran
Abstract
Background and Objectives:In any educational system, the optimal output of educational approach is of particular importance. Therefore, considering the personality characters of individuals and providing educational services in accordance with their characteristics are effective factors in learning and educational efficiency improvement. Analyzing the data related to learner’s behavior in an educational environment and implicitly discovering the learner’s personality based on their behavior is a well-noticed study in recent years. Over the last few years, using learners’ information such as number of friends, the level of activities in educational forum, writing style of learner, study duration, the difficulty of solved problem, the difficulty of presented example by learners, number of clicks, number of signs in sentences, the time spent doing homework are items that has been used to personal characteristic identification.
This study is aimed at using teammates’ changing / not changing data in order to learners’ personality identification. For this purpose the teammates’ changing/ not changing data extracted from a dynamic collaborative learning environment that allows leaners to change their teammate during the different sessions of learning, are used. The design and implementation of mentioned dynamic collaborative learning environment is based on game theory. Game theory provides mathematical models of conflict and collaboration between intelligent rational decision-makers.
Methods: In this paper, we collect teammates’ changing/not changing information of 119 randomly selected computer engineering students from a game theoretical dynamic collaborative learning environment. At the next step, using frequent pattern mining, as a tools of data mining, some aspects of the neo big 5 personality traits of learners are identified. In this survey, in order to evaluate the results, the extracted patterns from frequent pattern mining are compared with the neo big 5 personality questionnaire that have been filled by learners. In another part of research, using the Laplace’s rule of succession, valuable predictions were made about the probability of teammate’s changing of learners during the learning process.
Findings: In this study, using frequent pattern mining in learners’ behaviour, we identified some neo big 5 personality traits such as those in the first (neuroticism), second (extraversion), and third (openness to experience) dimensions, with an acceptable support value. The results of this part of research can be used in any adaptive learning environment that adapt learning process for individual learners with different personality. At the next step of our study, we predicted the probability of the teammate changing in the sessions after. At this step, we had a prediction accuracy of up to 67.44%. Using the results of this part, teammate suggestion can be made to learner based on likelihood of their teammates’ changing. That is, higher teammate changing probability, more appropriate teammate suggestion to learner.
Conclusion: The results of the present study can be used in any adaptive system that requires predicting group change behaviour or identifying personality dimensions based on behaviour.
Keywords
Main Subjects
COPYRIGHTS
©2020 The author(s). This is an open access article distributed under the terms of the Creative Commons Attribution (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, as long as the original authors and source are cited. No permission is required from the authors or the publishers.
', './data/jte/coversheet/cover_en.jpg'))
Send comment about this article