Data Preparation in Predictive Learning Analytics (PLA) for Student Dropout

Abstract

Predictive learning analytics (PLA) are the current trend to support learning processes. One of the main issues in education particularly in higher education (HE) is high numbers of dropout. There are little evidences being identified the variables contributing toward dropout during study period. The dropout are the major challenges of educational institutions as it concerns in the education cost and policy-making communities. The paper presents a data preparation process for student dropout in Duta Bangsa University. The number of students dropout in Duta Bangsa University are in high alarm for both management and also educator in Duta Bangsa. Preventing educational dropout are the major challenges to Duta Bangsa University. Data preparation is an important step in PLA processes, the main objective is to reduce noise and increase the accuracy and consistency of data before PLA executed. The data preparation on this paper consist of four steps: (1) Data Cleaning, (2) Data Integration, (3) Data Reduction, and (4) Data Transformation. The results of this study are accurate and consistent historical dropout data Duta Bangsa University. Furthermore, this paper highlights open challenges for future research in the area of PLA student dropout