Litcius/Paper detail

A Random Forest Students’ Performance Prediction (RFSPP) Model Based on Students’ Demographic Features

Saba Batool, Junaid Rashid, Muhammad Wasif Nisar, Jungeun Kim, Toqeer Mahmood, Amir Hussain

202130 citationsDOI

Abstract

Education plays a crucial role in individual life as well as for the whole nation. Many students are dropped out yearly in different academic courses. This study investigates the contribution of students’ demographic attributes to their academic achievements. The Random Forest classification model is used to predict students’ final exam performance. Three publically available datasets with a different set of demographic features are used to evaluate these attributes and their impact on students’ results. Hold-out and cross-validation methods are used to evaluate experimental results. Random Forest with three different datasets gave F-measure of 81.20%, 95.10%, and 84.16%. The presented study is significant for educational authorities to predict students’ performance before they drop out.

Topics & Concepts

Random forestComputer scienceSet (abstract data type)Mathematics educationDrop outMachine learningArtificial intelligencePsychologyDemographic economicsEconomicsProgramming languageOnline Learning and AnalyticsImbalanced Data Classification TechniquesMachine Learning and Data Classification
A Random Forest Students’ Performance Prediction (RFSPP) Model Based on Students’ Demographic Features | Litcius