Real Estate Price Prediction using Data Mining Techniques

The objective of this paper is to create a model for data mining using knowledge of real estate to predict property prices. It uses data set that have many dimensions to train and build a model. This is done in mainly two phases: data pre-processing and model building. In the first phase data is cleansed and normalized. Abnormal values are removed to make the data standardized. The second phase is mainly concerned with building various types of models to train and see which model performs the best and thus selecting the best model to use in the prediction of price for real estate properties. The models that have been used are elastic net, kernel ridge, lasso, random forest, SVM, XGBoost, LGBM and gradient boosting. Out of all these elastic net gave the highest accuracy score of 94%. Cross validation of models is done in phase two as well as methods to improve accuracy is discussed.