House Price Prediction Using Machine Learning

Abstract

India has experienced remarkable growth in recent years, spurred on by its high-tech sector, nice climate, and immigrant inflow. There is a significant demand for real estate properties as a result. This research article describes the creation of the house price prediction model, a machine learning-based tool, to meet this demand. A dataset is gathered from Kaggle that covers a variety of variables like area, rooms, location, and other amenities. It is trained and tested using a dataset of Indian real estate transactions. The models' accuracy is assessed, and the outcomes show that they perform at various levels. With a score of 64.5%, the SVR model has the lowest accuracy. The best model, however, is linear regression, which has a maximum accuracy of 84.5% compared to the other models. These results emphasize how important it is to choose the right algorithms for precise projections of property prices. Further research directions are suggested, such as the investigation of alternate data sources, the examination of extraneous aspects, and long-term forecasting. These options can enhance the precision and application of predictive models for predicting house prices and contribute to the growth and stability of the Indian real estate sector. The outcomes highlight linear regression's superior performance in this situation and offer stakeholders useful information for making strategic decisions.