Proposed classification for eLearning data analytics with MOA

Abstract

Elearning education has developed a crucial factor in the educational organization. With the situation of declining student size, elearning has to offer more cross-departmental and multi-disciplinary courses for individual needs to go over “one-size-fits-all” traditional model. Elearning data analytics which has not been professionally classified cannot produce reliable results. Classifications for elearning data help comfort the accuracy of outcomes and reducible pre-processing time. This research proposes a practical model for individual learning and personality. The proposed model based on data from the LMS classifies both the student preferences and personalities. The model helps design future curricula to suit student personalities, which intangibly assists them to be efficient in the study practice. Performance of the proposed classification is evaluated by using MOA software. It outperforms and improves the accuracy of complex elearning datasets. Besides, the results indicate an achievement in the students' study time after applying the association rule model on the elearning.