Predicting the Housing Market with Machine Learning

Abstract

The aim of this project is to build a machine learning model for predicting housing market prices using a dataset that includes information about MSSubClass, MSZoning, LotArea, LotConfig, BldgType, OverallCond, YearBuilt, YearRemodAdd, BsmtFinSF2, TotalBsmtSF and Sale Price. The dataset will be analyzed using exploratory data analysis (EDA) techniques to identify patterns and correlations between the different features and the housing prices. Several machine learning algorithms will be used to build the predictive model, including linear regression, SVR, Random Forest Regression, and CatBooster. The performance of the model will be evaluated using mean squared error and techniques such as hyperparameter tuning will be used to optimize the model's performance. The final model will be used to provide insights and predictions for future investment based on the price of a property in 5 years [1]. Key Terms – Correlation, Exploratory Data Analysis, Sale Price, Support Vector Regression.