Real Estate Price Prediction using Supervised Learning

Abstract

The least transparent sector of our economy is real estate. Housing prices change daily and are occasionally inflated rather than based on an appraisal. The central focus of our approach is using fundamental factors to forecast house values. Here, we strive to establish our assessments on each essential aspect when deciding the house's price. In our project, three elements affect a house's price: its physical attributes, design, and location. There have been a lot of studies utilizing typical machine learning techniques to estimate house prices effectively. Still, they need to pay more attention to how well each model performs and ignore the less well-known but more sophisticated models. Our project involves predictions using different Regression techniques like Linear Regression, Lasso Regression, and Decision Tree. Our project includes estimating the price of houses without any expectations of market prices and cost increments. The project aims to predict residential prices for customers considering their financial plans and needs. This project means to predict house prices in Pune city with various regression techniques. The project aims to predict cogent housing prices for those who do not own homes depending on their financial capabilities and desires. Estimating pricing will be possible by examining the mentioned goods, fare ranges, and advancements. This initiative aims to enable individuals to pinpoint the specific timeline for home acquisition and sellers in assessing the cost of a home sale. Spending resources on web-based apps without consulting a broker will benefit clients. Additionally, it provides a brief explanation of the various graphical and numerical techniques that are required to calculate the price of a home. Our study explains the goal of machine learning, the workings of the house pricing model, and the datasets that went into developing the model we suggest. Lasso, Decision Tree, and Linear Regression were among the models looked at in the study (accuracy: 83.54 percent) (accuracy -77.88 percent).