Predicting undergraduate students' perspectives on the E-learning obstacles with artificial neural networks

Abstract

The problem and the aim of the study. Educational institutions aim to provide an electronic learning environment that attracts students' interest and encourages them to exchange information, which considers eLearning a convenient way and means to develop. The rapid expansion in using eLearning might lead to obstacles during the teaching process. The present study aimed to investigate the eLearning obstacles from the undergraduate student's perspective at Al-Balqa Applied University through Artificial Neural Networks (ANN). Research methods. The researchers adopted two types of curricula, the descriptive and analytical approaches. The illustrative method is the definition of e-learning, neural networks and their fields of use and construction, and the application of a questionnaire to identify e-learning obstacles. The analytical approach applies the artificial neural network model to identify e-learning obstacles. Results. The result of the analyses indicated that there were different level degrees of four obstacle areas from the student's perspective: the most obstacle was faculty member obstacles (100%), followed by the infrastructure and technical support obstacles (95.4%), then university administration obstacles (81.1%), and the last one was the student's obstacles (80.3%). Also, the results showed differences in students' perspectives concerning their majors. Conclusion. The present study aimed to investigate the eLearning obstacles from the undergraduate student's perspective at Al-Balqa Applied University through Artificial Neural Networks, based on data collected from answers students on a questionnaire. The literature review indicated that neural networks outperform all other classifiers in prediction accuracy. A multi-layer neural network has been trained through a backpropagation algorithm to predict e-learning obstacles. The accuracy rate of the classification was very high. The results show the differences in students' expected predictions for e-learning obstacles according to the student major. The order of the student predictions for eLearning obstacles from most important to least important was as follows: faculty member obstacles, faculty member obstacles, infrastructure and technical support, university administration and the student's barriers. The reality of the application of e-learning showed that undergraduate students of Princess Alia University College had a negative perspective on eLearning. The study recommended improving the capability of the students' academic advisors in employing e-learning through training to improve their proficiency in using e-learning.