Real Estate Search and Valuation to get best valued sites using Linear Regression Algorithm and Compared with Random Forest Algorithm.

Abstract

Aim: The main aim of this study is to forecast Real Estate Prices using Novel Linear Regression and Random Forest algorithms. Accordingly, A prediction model based on Novel Linear Regression  is proposed for prediction of Housing Price as well as in the process of features selection. There are certain factors that influence the price of the houses which includes location, Conditions, Age, square area etc. Compared with other methods, our work can obtain better performance through different experiments. Materials and Methods: Linear Regression and Random Forest algorithms are used to predict the Real Estate Prices. Sample size is calculated using G power calculator and found to be 25 per group has been taken and a total of 50 samples are used.  Pretest power is 80% and CI of 95% and the significance value is (p<0.05). Results: Based on the study Linear Regression has significantly more accuracy (85%) compared with Random Forest algorithm (78%) and the significance value p=0.01 (p<0.05). It shows that there is a significant difference between two groups. Conclusion: According to this study Linear Regression has better accuracy than the Random Forest algorithm to predict the Real Estate Prices.