Boston Housing Price Prediction

A comprehensive end-to-end data science and machine learning project designed to analyze real estate data and accurately predict house prices using the benchmark Boston Housing dataset.

Key Features & Highlights:

Exploratory Data Analysis (EDA): Cleaned data, analyzed feature correlations, and handled missing values to ensure high model efficiency.

Model Training: Trained robust predictive machine learning algorithms and serialized the final model using Joblib.

Interactive Web Application: Developed an intuitive web interface (app.py) allowing users to input property specifications and instantly receive real-time price predictions.

Clean Project Structure: Organized following software engineering standards, including training scripts (train.py), web app script (app.py), requirements specification (requirements.txt), and comprehensive documentation (README.md).

Technologies Used:

Programming Language: Python

Data Science & ML Libraries: Scikit-Learn, Pandas, NumPy

Web App Framework: Streamlit / Flask

Model Serialization: Joblib

--------------------------------------------------------------------------------------------------------------------------------------------------------

boston_project/

├── housing.csv # Dataset used for training

├── model.joblib # Serialized trained Decision Tree model

├── metrics.json # Model evaluation metrics and feature boundaries

├── train.py # Model training and hyperparameter tuning script

├── app.py # Streamlit web application

├── requirements.txt # Python dependencies

└── notebook/

├── boston_housing.ipynb # Exploratory Data Analysis & experiments

└── visuals.py # Visualization helpers