Regression and Classification with the Ames Housing Data

Concept

This project is focused on:

1. Develop an algorithm to reliably estimate the value of residential houses based on fixed characteristics.
2. Identify characteristics of houses that the company can cost‑effectively change/renovate with their construction team.
3. Evaluate the mean dollar value of different renovations.

Data

The dataset was obtained from the publicly available Ames housing data recently made available on Kaggle.

Approach

Perform any cleaning, feature engineering, and EDA deemed necessary. Remove any houses that are not residential from the dataset. Identify fixed features that can predict price. Train a model on pre‑2010 data and evaluate its performance on the 2010 houses. Characterize the model: how well does it perform, and what are the best estimates of price?

Algorithms Used

In order to analyze the data and generate insights out of it, the process looked at the data using a range of visualizations, then applied the following models: