Predicting House Rental Value Among Students in Higher Institution Using Data Mining Techniques

Abstract

Rental housing value is a prominent issue among students in higher institutions especially those who stay off–campus as the stay off-campus figure has been increasing these past years. The aim of this research is to determine the students’ perception on the level of agreement on potential factors (agreement on rental fee, socializing, housing facilities, privacy) that might affect the
rental housing value. This study also examined the characteristics that determine the rental house value among university students. This study was conducted on 377 selected students from UiTM, UMK and USM in Kelantan, Malaysia. Stratified sampling was used to select the samples and cross-sectional study design was employed. The data was analysed using data mining
approach of ID3 Decision Tree, CART Decision Tree and Neural Network classifier. The study found that facilities indicate the highest score of students’ perception on the level of agreement of potential factors affecting rental housing value. Based on the predictive model, ID3 Decision Tree is the best model in determining the factors that significantly influence the rental value. The
most important variable in determining the rental house value is whether the house is equipped with these three important facilities (microwave, single room and television) or not.