Used Car Prices

This repository contains a complete data science workflow for predicting used car prices using machine learning.

🔗 Original Notebook on Kaggle: View here
📓 Notebook on GitHub: View here

📈 Performance

• Model: LightGBM Regressor
• Approach: Gradient Boosting with feature engineering and target transformation

🛠️ Key Features

• Data Cleaning:

  • Handling missing values
  • Standardization of text features
  • Extraction of structured information from raw data

• Feature Engineering:

  • Creation of new features such as age and hp
  • Log transformation of skewed variables
  • Encoding of categorical features

• Target Transformation:

  • Log transformation (log1p) applied to price
  • Inverse transformation using expm1

• Modeling:

  • LightGBM implementation for tabular data
  • Early stopping to prevent overfitting
  • Hyperparameter tuning

🚀 How to use

1. Clone the repo: git clone https://github.com/lucalullo/Used-car-prices.git
2. Install dependencies: pip install pandas numpy scikit-learn lightgbm matplotlib seaborn
3. Run the used-car-prices.ipynb notebook

---

Author: Luca Lullo
Data Scientist | Machine Learning Applied