The Impact of Data Normalization on Predicting Student Performance: A Case Study from Hashemite University
Year: 2020
Type: article
Abstract: The main goal of mining and analyzing educational data is to identify interesting patterns that might influence learning quality and student performance.Supervised machine learning can be utilized to predict student performance.More specifically, classification algorithms are used to construct a student performance prediction model.One of the main challenges to generate the desired model is adopting appropriate data preprocessing techniques.The central idea of the work presented in this paper is to generate a student performance prediction model and investigate the impact of various data normalization techniques on its effectiveness.The influence of various data normalization techniques is assessed on three novel academic datasets collected from the Hashemite University in Jordan.The obtained results provided empirical evidence that the adopted data normalization technique has a significant effect on classification model effectiveness.The SVM classifier coupled with the Z-score normalization technique produced the most effective student performance prediction model.
Author: Esra'a Alshdaifat
Institution: Hashemite University
Language: English
Subfield: Computer Science Applications
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: bronze
Funder: Hashemite University
OpenAlex ID: W3086104386