Predicting Housing Prices
Abstract
In this project, price of houses in Chicago, Illinois region are predicted. First, data pertaining to specific characteristics of the house is gathered. This is combined with data from other sources, such as, socioeconomic conditions of the regions, crime-rates in different localities and distance from nearest CTA ’L’ station. Exploratory analysis is then performed to identify trends, correlation and derive other interesting insights from the data. Following this, machine learning models are built to accurately predict the price of a house, and the performance of the models are measured. Model selection is performed based on the performance metric to identify the best performing model and validate the accuracy of the predictions on new data. We conclude that tree-based XGBoost model predicts the house price most accurately. We also observe some important predictor variables, including square footage of the house, the number of baths in the house and the socio-economic condition of the communities.
If you are interested in more details, please refer to a full report