House Prices

This repository contains a comprehensive data science workflow for predicting residential home prices.

🔗 Original Notebook on Kaggle: View here

📈 Performance

• Kaggle Score (RMSE): 0.12676
• Model: Ridge Regression with hyperparameter tuning.

🛠️ Key Features

• Exploratory Data Analysis (EDA): Detailed visualization of target distribution and correlation matrices.
• Advanced Preprocessing: - Handling missing values based on feature context.
  • Box-Cox transformation and Log-scaling for skewed numerical features.
  • Categorical encoding for quality-related features.
• Modeling: Implementation of Regularized Linear Models (Ridge) to prevent overfitting given the high number of features (79+).

🚀 How to use

1. Clone the repo: git clone https://github.com/lucalullo/House-price.git
2. Install dependencies: pip install pandas numpy seaborn matplotlib scikit-learn scipy
3. Run the house-prices.ipynb notebook.

---

Author: Luca Lullo
Data Scientist | Machine Learning Applied