<![CDATA[The need for a place to live is one that many people prepare, both millennials and adults and the elderly. With the continued increase in population growth in Indonesia and increasing public interest in buying a place to live early on, this can make not all groups of people have a place to live or a house that is quite livable. Related to this, the public needs up-to-date information related to predictions of house prices both for housing and second-hand housing prices for planning purposes in the future. The purpose of this study is to carry out a comparative analysis of the prediction results of house prices with several Machine Learning algorithms consist of Linear Regression, Random Forest Regression and Gradient Boosted Trees Regression. Evaluation for all the method applying Cross-Validation. The evaluation is seen from the smallest Root Mean Square Error (RMSE) error rate of each testing method. The results of this study are the Random Forest Regression obtained an RMSE value of 0.440, the Linear Regression model obtained an RMSE value of 0.515 and the RMSE value of Gradient Boosted Trees Regression of 0.508. The results were obtained from testing a dataset of 2011 records with a division of 80% for data training and 20% for data testing, the data has 6 attributes used in testing including house prices, land area, building area, number of bathrooms, number of bedrooms and the number of garages. In this study, prediction results using the Random Forest Regression method yielded the highest accuracy of 81.5% compared to the Linear Regression and Gradient Boosted Trees Regression methods.]]>
Copyrights © 2023
