Forecasting Selling Price For Housing

The Saratoga Country of New York in-demand place to live, work and visit. Their cultural venues, excellent schools, and colleges along with trails, parks, and recreation programs make the country a perfect place to live. Forecasting the fluctuating rental price and future demand for the same rental space is a crucial task.

In future developers need a system to predict the house price, to determine the selling price of a house, and the customer to arrange finances at the right time for the purchase of the house. Few factors that influence the price of houses like the location, physical conditions, etc.

In this use case, we aim at predicting the house prices of Saratoga Country -New York, by leveraging regression models in Machine Learning and other Data Science tools and techniques to obtain optimal prediction values. Apart from determining the prevailing price of the land, the systems also aim at determining the ease of access to other public facilities like educational institutions, hospitals, recreational facilities, etc.

Through this research, we have accomplished to build a machine-learning-based Predictive analytical model. The model is completely built on adopting the regression models to analyze the available data. Using this model we could predict the price fluctuations of the houses in Saratoga country.

We use regression models of machine learning models to understand which model has been predicted with the utmost accuracy.

• 1. Linear regression
• 2. Lasso regression
• 3. Ridge regression
• 4. Random forest

The level of accuracy captured after each model is linear regression-65.88%; random forests-65.09%; ridge regression-65.82%; lasso regression-65.88%.