7006SCN_Varsha_16948822

NYC Taxi High-Tip Prediction Using PySpark

This project applies Machine Learning and Big Data techniques to predict whether an NYC Yellow Taxi trip will receive a high tip (15% or more). Using over 10 million taxi trip records, Apache Spark was employed for distributed data processing, feature engineering, and scalable model training. Four classification algorithms—Logistic Regression, Decision Tree, Random Forest, and Gradient Boosted Trees (GBT)—were evaluated and compared using Accuracy, Precision, Recall, F1-Score, and ROC-AUC metrics. Results showed that GBT achieved the best overall performance. Tableau dashboards were developed to visualize data quality, model performance, customer tipping behavior, and scalability insights, supporting data-driven business decision-making.

7006SCN2526MAYJUL/7006SCN_Varsha_16948822

About

Resources

Stars

Watchers

Forks

Releases

Languages