SMART HOUSE PRICE VALUATION USING AI-BASED PREDICTIVE TECHNIQUES

Abstract

A variety of machine learning algorithms were implemented in our research to forecast the sale prices of properties. The ultimate selling price of a house is influenced by a variety of factors, including its size, neighborhood, age, construction style, number of bedrooms, garage space, and more. In this paper, we implement machine learning algorithms to construct a home prediction model. A prediction model is developed by employing machine learning techniques such as logistic regression, decision trees, support vector regression, and Lasso regression. The housing statistics of three thousand residences were examined. There are four models: Logistic Regression, SVM, Lasso Regression, and Decision Tree, each with an R-squared value of 0.98, 0.96, 0.81, and 0.99. These algorithms were evaluated using a variety of metrics, such as accuracy, RMSE, MSE, and MAE. This research serves as an additional illustration of the significance of our methodology and approach.