BACKGROUND
This paper investigates the use of mobile learning platforms for learning purposes among university students in UAE. An extended Technology Acceptance Model (TAM) and theory of planned behavior (TPB) are proposed to analyze the adoption of mobile learning platforms by university students for accessing course materials, searching the web for information related to their discipline, sharing knowledge, conducting assignments during COVID-19 pandemic. The total number of questionnaires collected was 1880 form different universities. Partial least squares-structural equation modeling (PLS-SEM) and machine learning algorithms (ML) were utilized to investigate the research model based on the student’s data gathered through a survey. According to the results, each hypothesized relationship within the research model has been supported by the data analysis methods. It should also be noted that the J48 classifier mostly had the upper hand on other classifiers when it comes to the prediction of the dependent variable. As per the indication of our research, teaching and learning can greatly benefit from the adoption of machine learning as an educational tool at the time of this pandemic; nevertheless, its significance could be lowered because of the emotion of fear concerning poor grades, stressful family circumstances, and loss of friends. Accordingly, this issue can only be solved by evaluating the emotions of students during this pandemic.
OBJECTIVE
This study is one of the earliest attempt to: (1) theoretically integrate the notion of fear within a hybrid model of Technology Acceptance Model (TAM) & Theory of Planned Behavior (TPB) (2) empirically test the effect of COVID-19 on the users of mobile application, and (3) explore the impact of the Coronavirus pandemic on users' ability to use the mobile application easily and users' attitude towards the usefulness of mobile learning platform.
METHODS
The developed theoretical model has been evaluated using two different techniques in this research. The first one involves the usage of the partial least squares-structural equation modeling (PLS-SEM) alongside the SmartPLS tool. This research uses PLS-SEM mainly because both the structural and measurement model can be concurrently analyzed through PLS-SEM, which increases the preciseness of results. As for the second technique, the research predicts the dependent variables entailing the conceptual model with the help of machine learning algorithms via Weka.
RESULTS
The present research has implemented a model that would be useful for future studies to be conducted since it helps assess the COVID-19 influence at the time of the pandemic period. Keeping the research results in mind, and the fear factor present during the period, the ML is considered to be a significantly useful tool which helps reduce the fear present within the peers and instructors. Similarly, the perceived fear (PF) highly affects the PU and PEU. According to the responses, during the pandemic period, the PF is quite evident; however, the ML maintains a high PU and PEU degree, which reduces the fear factor and encourages the students to participate in their scheduled class.
CONCLUSIONS
The current research results are similar to the ones presented in earlier research studies related to the TAM and TPB variable’s importance (Ajzen, 1985; F. D Davis, 1989; Teo, 2012; V Venkatesh & Bala, 2008). It is observed that the students are much more acceptable towards technology is there is nothing but the ML technology available as the tool for learning during the COVID-19 pandemic. The PU and PEU related results are also similar to the ones of the earlier PU and PEU related results that influence the student acceptance of ML. Hence, it should be considered as an indicator for the students intention to make use of the ML when the environment is infected with COVID-19. Furthermore, PU is highly affected by PEU, which indicates that if it is easy to use the technology, then it would be considered useful.
Predicting the intention to use mobile learning during Coronavirus Pandemic through Machine Learning Algorithms (Preprint)
