Big Data Analytics: Student Performance Prediction Using Feature Selection and Machine Learning on Microsoft Azure Platform

Authors

  • Wattana Punlumjeak Department of Computer Engineering,Faculty of Engineering, Rajamangala University of Technology Thanyaburi, Pathumthani, Thailand.
  • Nachirat Rachburee Department of Computer Engineering,Faculty of Engineering, Rajamangala University of Technology Thanyaburi, Pathumthani, Thailand.
  • Jedsada Arunrerk Department of Computer Engineering,Faculty of Engineering, Rajamangala University of Technology Thanyaburi, Pathumthani, Thailand.

Keywords:

Big Data Analytics, Feature Selection, Microsoft Azure, Student Prediction,

Abstract

In recent years, big data analytics has been a new growing research area and the essence of cloud computing is used to support a shared pool of resources. In educational mining, the huge volume of student data needs analytics technologies to extract valuable knowledge. It has been recognized that a high performance accuracy of student prediction model will be helpful for student and stakeholders. In this experiment, feature selection methods were proposed to identify the most significant and intrinsic features before classification methods were used. Experiment was conducted to evaluate the performance of the prediction model. The result of the experiment showed that mutual information in feature selection method with neural network classifier gave the best overall accuracy at 90.60% for student’s data at Rajamangala University of Technology Thanyaburi. This experiment is extremely useful for students, teachers and management to find useful knowledge not only in identifying the problem areas and reasons that affect student’s performance, but also in understanding the feature selection and classification methods, which are the most effective way to analyze student’s performance on a cloud computing environment.

Downloads

Published

2017-03-15

How to Cite

Punlumjeak, W., Rachburee, N., & Arunrerk, J. (2017). Big Data Analytics: Student Performance Prediction Using Feature Selection and Machine Learning on Microsoft Azure Platform. Journal of Telecommunication, Electronic and Computer Engineering (JTEC), 9(1-4), 113–117. Retrieved from https://jtec.utem.edu.my/jtec/article/view/1791