Determination of Mobile Learning Application Usability: A Pattern Mining Approach that Goes Beyond the Statistical Method

Abstract

Use of mobile learning applications in student-centered learning is rapidly growing in the present globe. Student satisfaction needs to be significantly considered to provide the maximum academic outcome on their learning. Hence, evaluating mobile learning systems (MLS) to test their usability is important. The different types of statistical approaches have been used to test the usability of previous MLS. The main objectives of this study are to evaluate the usability of the mobile learning system using a data science approach and also making a comparison with a statistical approach. To evaluate the proposed mobile learning system through the data science approach, questionnaire responses were obtained from 100 learners. These responses were evaluated using two pattern mining algorithms namely Apriori and FP-Growth. According to the results, the Apriori algorithm shows 94% system usability while the FP-Growth algorithm ensures 93% system usability. According to the statistical approach, the overall mean value 4.083 of the questionnaire responses was obtained as the usability of the system. Finally, it is concluded that the pattern mining approach is more noticeable than the statistical method comparatively when determining the usability of MLS.