Accepted to AI for Urban Planning @ AAAI 2025
Authors: Joshua Peguero, Felix Lee
Accurate ridership prediction is essential for efficient public transit planning and resource allocation. We present a neural network framework for predicting ridership across New York City's transit system, incorporating temporal patterns, weather data, and special events. Our model achieves significant improvements over baseline methods and provides interpretable insights into factors affecting ridership, supporting data-driven decision making for urban transportation planners. [arXiv link TBA]
