House Price Prediction using regression techniques

Abstract

Real estate companies keep special care when they purchase or sell a new house. A significant amount of expertise and market awareness is required for making accurate price predictions of houses so as to turn it out into a profitable investment. Also, the setting of prices manually is quite difficult and tedious task and the accuracy of prediction done by the real estate experts is also not good in that case due to the probability of human errors. The primary goal of this project is to provide the best and most profitable house price predictions to real estate investors with the help of a machine learning model so that they can get best returns in their deals as per the market scenario. It can also provide a lot of convenience to buyers who want to purchase a house at best market prices depending upon the desired house features. It consists of a ML-based model which mainly involves the use of three machine learning algorithms namely linear regression, decision tree regression and random forest regression. Three individual models are created on the basis of each of these algorithms and the most suitable model with maximum accuracy is chosen. With the analysis of various features and specifications of the houses such as number of rooms, locality, population status, physical conditions etc., prices will be estimated.