Machine learning techniques to predict housing prices and identify factors affecting them

• Author(s) / Creator(s)

  • Patel, Aditya Kalpeshbhai
  • Mirza, Fawwad Ahmed

• House prices have a big impact on the economy, and customers and real estate agents are quite concerned about the price changes. Every year, housing prices rise, which ultimately highlights the necessity for a method or plan that might forecast home prices in the future. According to research, house owners and the real estate industry frequently worry about price swings in real estate. To identify pertinent features and the most effective models to anticipate home values, a review of the literature is conducted before the creation and analysis of machine learning models. Physical attributes including the living area, plot area, location, number of bedrooms etc. affect the price of a property directly. Previous studies have been based on just a few of these variables. This paper used dataset for USA housing with different internal characteristics and applied CART with ensemble techniques (Boosting and Bagging), K-Nearest Neighbor, Local Regression and Random Forest to predict property prices of houses and their relationship with different features. The paper will validate the different machine learning techniques applied by using k-fold cross validation and RMSE to provide an optimistic view on property price prediction.

• Date created

  2023

• Subjects / Keywords

  • propery price prediction
  • real estate forecast
  • classification and regression trees
  • CART
  • k-nearest neighbors
  • random forest
  • house attributes
  • cross validation

• Type of Item

  Research Material

• DOI

  https://doi.org/10.7939/r3-z6ys-t488

• License

  Attribution-NonCommercial 4.0 International