House Price Prediction in Southern Chennai Using Machine Learning Algorithms

The housing market in India, particularly in urban areas, has experienced a significant surge in prices. Accurately determining the value of houses is crucial for informed decision-making by investors, realtors, banks, and government agencies. This study aims to utilize machine learning techniques, specifically the K-nearest neighbor (KNN), linear regression, and random forest algorithms, to predict house prices in Chennai, India. Numerous factors that influence the house prices, including amenities and proximity to surrounding facilities, were considered for the study. A dataset comprising 330 houses and twenty-nine factors was collected for analysis. The models were trained using 70% of the collected data, while the remaining 30% were utilized for testing. By employing the developed models, house prices were predicted, and their accuracies were compared. The mean squared error (MSE), root mean squared error (RMSE), and mean absolute error (MAE) were employed as performance evaluation metrics. Notably, the random forest algorithm outperformed both the K-nearest neighbor and linear regression methods in terms of accuracy of predicting the house prices.