Early and dynamic student achievement prediction in e-learning courses using neural networks

Citation

(2009), "Early and dynamic student achievement prediction in e-learning courses using neural networks", Development and Learning in Organizations, Vol. 23 No. 4. https://doi.org/10.1108/dlo.2009.08123dad.006

Early and dynamic student achievement prediction in e-learning courses using neural networks

Article Type: Abstracts From: Development and Learning in Organizations, Volume 23, Issue 4

Lykourentzou I., Giannoukos I., Mpardis G., Nikolopoulos V., Loumos V. Journal of the American Society for Information Science and Technology, February 2009, Vol. 60 No. 2, Start page: 372, No. of pages: 9

Purpose – To develop feed-forward neural networks for use in e-learning systems designed to predict gradually the final grades of students who took an introductory level e-learning course. Design/methodology/approach – A review of the e-learning literature is presented. Describes the development off the student achievement prediction method, aimed at being applied to a ten-week introductory level e-learning course, based on multiple feed-forward neural networks for the dynamic prediction of students’ final achievement and the clustering of the student data in two virtual groups according to their performance. Reports the results of experiments in which the technique was applied to multiple-choice test grades as the input data set relating to 57 students who completed the Spring 2006 and Spring 2007 courses. Compares the results with those obtained using linear regression. Findings – The results indicated that the technique allowed accurate prediction to be made at an early stage (third week of the ten-week course) and the adequacy of the approach was further demonstrated by the low misplacement rates when students were clustered. Notes that the comparison with linear regression revealed the neural network approach to be more effective in all prediction stages. Originality/value – Provides a prediction method that can easily be integrated into an e-learning management system and promote greater interactivity. ISSN: 1532-2882 Reference: 38AG545

Keywords: Computer based learning, Computer-based training, Neural nets, Software tools, Students, Teaching aids