Robust Taxi Fare Prediction Under Noisy Conditions: A Comparative Study of GAT, TimesNet, and XGBoost
This work addresses robust fare prediction for ride-hailing platforms, but it is incremental as it compares existing methods on a new dataset.
This study tackled the problem of predicting taxi fares under noisy conditions by comparing Graph Attention Networks (GAT), XGBoost, and TimesNet on a dataset of over 55 million records, finding critical differences in performance between classical and deep learning models.
Precise fare prediction is crucial in ride-hailing platforms and urban mobility systems. This study examines three machine learning models-Graph Attention Networks (GAT), XGBoost, and TimesNet to evaluate their predictive capabilities for taxi fares using a real-world dataset comprising over 55 million records. Both raw (noisy) and denoised versions of the dataset are analyzed to assess the impact of data quality on model performance. The study evaluated the models along multiple axes, including predictive accuracy, calibration, uncertainty estimation, out-of-distribution (OOD) robustness, and feature sensitivity. We also explore pre-processing strategies, including KNN imputation, Gaussian noise injection, and autoencoder-based denoising. The study reveals critical differences between classical and deep learning models under realistic conditions, offering practical guidelines for building robust and scalable models in urban fare prediction systems.