🏡 Boston Housing Price Prediction AI

This is an interactive web application for predicting Boston housing prices using advanced Machine Learning models. Built with Python and Streamlit.

🌟 Features

• AutoML Leaderboard: Compare Linear Regression, Random Forest, XGBoost, LightGBM, and CatBoost performance instantly.
• Explainable AI (XAI): Integrated SHAP and LIME for model transparency. Understand why the model made a specific prediction.
• What-If Simulator: Interactive sidebar to adjust house features and see real-time price predictions.
• Deep Learning: Includes a PyTorch Neural Network demo.

🏠 Input Features Explained (Sidebar Variables)

Use the sidebar to adjust these parameters and simulate different housing conditions:

Feature	Description	Impact on Price (General)
CRIM	Per capita crime rate by town.	📉 Higher crime → Lower price
ZN	Proportion of residential land zoned for lots over 25,000 sq.ft.	📈 More residential zone → Higher price
INDUS	Proportion of non-retail business acres per town.	📉 More industry → Lower price
CHAS	Charles River dummy variable (1 if tract bounds river; 0 otherwise).	📈 River view → Higher price
NOX	Nitric oxides concentration (parts per 10 million).	📉 High pollution → Lower price
RM	Average number of rooms per dwelling.	📈 More rooms → Higher price
AGE	Proportion of owner-occupied units built prior to 1940.	📉 Older buildings → Lower price
DIS	Weighted distances to five Boston employment centres.	📈 Closer to centers is usually better, but DIS measures distance to work.
RAD	Index of accessibility to radial highways.	📈 Better transport access → Higher price
TAX	Full-value property-tax rate per $10,000.	📉 Higher tax → Lower price
PTRATIO	Pupil-teacher ratio by town.	📉 More students per teacher → Lower price
LSTAT	% lower status of the population.	📉 Higher % lower status → Lower price

🚀 How to Run Locally

If you want to run this app on your own machine:

1. Clone the repository:

   git clone https://github.com/2002brian/boston-housing-app.git
   cd boston-housing-app
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Run the app:

   streamlit run app.py
   
   

📊 Understanding the Evaluation Metrics

The dashboard displays two key metrics to evaluate model performance on the test set:

1. R-squared (R²)

• What it is: A statistical measure representing the proportion of variance for a dependent variable explained by the independent variables.

• Interpretation:

• Value is between 0 and 1.

• 1 means perfect prediction.

• 0 means the model fails to explain any variability.

• Goal: Higher is better.

2. Root Mean Squared Error (RMSE)

• What it is: The standard deviation of the residuals (prediction errors). It tells you how concentrated the data is around the line of best fit.

• Interpretation:

• Measured in the same units as the target variable ($1000s).

• 0 would mean perfect predictions.

• Goal: Lower is better.

Created by Brian Wu