Performance evaluation of radial basis function (RBF) and multilayer perceptron (MLP) in identifying the contributory factors of learning habits during COVID-19

• Published in: AIP Conference Proceedings
• Main Author: Mangshor N.N.A.; Ibrahim S.; Sabri N.; Hamzah H.H.M.; Kamaruddin S.A.
• Format: Conference paper
• Language: English
• Published: American Institute of Physics 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85189289576&doi=10.1063%2f5.0150371&partnerID=40&md5=6f4289c0d14108f3e1bfb499098c45ed

The Coronavirus Disease 2019 (or known as COVID-19) outbreak has affected a lot of people worldwide and resulted in the practice of new norms in most daily activities. One of the practices that is closely associated with the new norm is online learning which has also become a necessity and essential especially in the education sector. This has led to the need to investigate the contributory factors of students' learning habits as students' learning habits correlate with students' performance. A dataset from Vietnamese students' learning habits during the COVID-19 pandemic is used where the performance of the Radial Basis Function (RBF) network and Multilayer Perceptron (MLP) network are compared. There are fifteen covariates used in this study which are fa_job, exam, Self_evaluation, English, Lh_before_Cov, nec_prog, nec_habit, nec_parent, eff_moti, eff_con eff_supp, eff_env, eff_obj, eff_resource, and eff_friend. Based on the comparison performed, it is concluded that both RBF and MLP networks are capable in identifying the contributory factors of students' learning habits during the COVID-19 pandemic. It is monitored that the MLP network outperforms the RBF network with the lower Sum of Squares Error (SSE) and Relative Error (RE) values. Additionally, it is also observed that the learning hours before COVID-19 (Lh_before_Cov) is identified as the most contributing factor in students' learning habits. © 2024 AIP Publishing LLC.