Residential Property Price Prediction Using Machine Learning: MakanSETU

Abstract

MakanSETU is an emerging and advanced solution in the Real Estate industry. Real Estate Industry is at boom in the 21st century and trading Real Estate has become a great opportunity for Real Estate owners as well as others. The projection of Real Estate industry in business acquisitions is expected to reach 11 trillion USD. However, there is no proper solution to deal with inaccurate prices of properties online. The system proposed in this paper uses Native and new age Machine learning algorithms to predict and validate value of residential properties. Supervised learning is used in the system along with multiple Regressors to obtain the best result. Some of the regression algorithms used are Simple Linear regression, Decision tree regression, Random Forest regression (100 n-trees, 200 n-trees, and 500 n-trees), and Extreme Gradient Boost regression algorithm. The development of this system has followed a series of Data Collection, data handling, data processing, EDA, Feature engineering and Feature selection. The system enables investors to get a fair value of a property. The system is considered successful and ready to implement in the real work.