Machine Learning Project: Transportation Metrics Prediction
A project predicting transportation metrics using machine learning algorithms such as Linear Regression, Random Forest, and LightGBM. LightGBM achieved the best performance with an MAE of 5.534 and an R-squared value of 0.861.
In this project, I developed a predictive model to estimate transportation metrics, exploring various machine learning algorithms such as Linear Regression, Random Forest, and LightGBM. LightGBM emerged as the optimal model due to its superior performance in terms of accuracy, efficiency, and scalability.
Extensive feature engineering was conducted, including the creation of interaction features like speed. The LightGBM model achieved the lowest Mean Absolute Error (MAE) of 5.534 and the highest R-squared value of 0.861.
Additionally, cross-validation confirmed its robustness, making it well-suited for real-world applications. Through this project, I gained valuable insights into data preprocessing, model tuning, and performance evaluation.
Technologies: Pandas, NumPy, scikit-learn, LightGBM, Seaborn
