There are numerous housing options available in today's society, so anyone can choose an affordable home that is close to nature and equipped with all the necessary amenities. However, the rising cost of housing has made it difficult to construct new housing types economically for the public benefit. Rather than estimating a single number, it is sometimes more useful and enticing to forecast a range of property value declines. Due to the difficulty in categorizing products, it may be difficult to estimate pricing. Academics frequently use the House Price Index (HPI) query method to attempt accurate forecasts of future house price fluctuations by monitoring average price changes across multiple acquisitions or refinancing operations involving identical properties. The fact that real estate price trends are affected by multiple variables, such as population density and geographic location, adds to the complexity of this issue. In this study, we use four algorithms to estimate and propose housing prices: linear regression, a decision tree, a random forest regressor, and a gradient boost regressor. The purpose of these steps is to develop a dependable machine-learning model for use in predictive analytics and data classification. We have also contrasted the accuracy of each individual's predictions. The research utilizes the Bangalore dataset, which contains more than 13,000 instances. Due to the high quality of its results, Gradient Boosting is an efficient method for recommending secure real estate investments. The Gradient Boost Regressor, the most accurate of these options, was chosen as the primary model.