fbpx Aligning Learning Materials and Assessment with Course Learning Outcomes in MOOCs Using Data Mining Techniques |ARAB AMERICAN UNIVERSITY
Contact information for Technical Support and Student Assistance ... Click here

Aligning Learning Materials and Assessment with Course Learning Outcomes in MOOCs Using Data Mining Techniques

Chapter Number: 
1
Authors: 
Ahmad Ewais
Mohammed Awad
Khetam Hadia
Pages From: 
1
To: 
25
Book Title: 
Advances in Integrations of Intelligent Methods
Editor(s): 
Hatzilygeroudis I., Perikos I., Grivokostopoulou F.
Publisher: 
Smart Innovation, Systems and Technologies,
ISBN: 
978-981-15-1917-8
Date: 
Saturday, January 18, 2020
Abstract: 
Massive open online courses (MOOCs) are considered as a new trend in the domain of e-learning. They provide a platform for supporting learners from different places and at any time with highly scalable and interesting learning experience. As a result, this led to continuous increase in the rate of learners with different knowledge, background, and skills. Therefore, supporting learners with adapted courses’ materials and assessments based on learning outcomes is considered as a crucial concept for enhancing MOOCs. This paper presents a framework for delivering learning materials and generating assessments based on intended learning outcomes (ILOs) using both support vector machine (SVM) and fuzzy logic algorithm. The SVM is mainly used to classify learning materials according to learning outcomes. On the other hand, the fuzzy logic algorithm is mainly used to generate examinations and quizzes automatically based on learners’ achievements and scores. Accordingly, the proposed framework can be used to enable learners to achieve learning outcomes by following adapted learning materials and automatically generated examinations. To validate the proposed framework, a prototype was developed and evaluated. The results of classifying both learning materials and assessment show interesting results. For instance, the results of classification process for learning materials were related to a number of factors such as accuracy rate for SVM classifier which was 71.5%, the weighted average for TP rate = 0.715, FN rate = 0.285, FP rate = 0.028, TN rate = 0.972, precision = 0.738, recall = 0.715, F-measure = 0.678, and finally ROC area = 0.939. In addition, fuzzy logic technique provided promising results to deliver examinations with a difficulty levels that are compatible with the current level of the learner depending on his grade point average (GPA).