"Decision Tree Regressor Compared with Random Forest Regressor for House Price Prediction in Mumbai"

Abstract

Aim: The aim of the work is to evaluate the accuracy in predicting house price in Mumbai Using Decision Tree Regressor and comparison with Random Forest Regressor. Materials and Methods: The work was carried out on 1259 records taken from a Mumbai house dataset. A framework for house price prediction in the real estate sector comparing Decision Tree Regressor and Random Forest Regressor has been proposed and developed. The sample Size was calculated as 188 in each group using G power value is 0.8. Sample Size was calculated using clinical analysis, with alpha and beta values 0.05 and 0.2, Pretest power 80% with enrollment ratio 1 and significance(p) is 0.038. Results: The Decision Tree Regressor  produces 71.63% accuracy  in predicting house prices on the Mumbai dataset, whereas the Random Forest Regression model produces 62.11% accuracy and with a significance score of 0.03 (p>0.05), there is a significant difference between two groups. Decision Tree Regressor is better than Random Forest Regression. Conclusion: The results show that the performance of Decision Tree Regressor is better when compared with Random Forest Regression in terms of accuracy.