Artificial Neural Networks Applied to Taxi Destination Prediction

Abstract

We describe our first-place solution to the ECML/PKDD discovery challenge on taxi destination prediction. The task consisted in predicting the destination of a taxi based on the beginning of its trajectory, represented as a variable-length sequence of GPS points, and diverse associated meta-information, such as the departure time, the driver id and client information. Contrary to most published competitor approaches, we used an almost fully automated approach based on neural networks and we ranked first out of 381 teams. The architectures we tried use multi-layer perceptrons, bidirectional recurrent neural networks and models inspired from recently introduced memory networks. Our approach could easily be adapted to other applications in which the goal is to predict a fixed-length output from a variable-length sequence.

ECML/PKDD discovery challenge

References in corpus (1)

• Blocks and Fuel: Frameworks for deep learning

Cited by in corpus (11)

• fastai: A Layered API for Deep Learning
• DeepTravel: a Neural Network Based Travel Time Estimation Model with Auxiliary Supervision
• Improving Missing Data Imputation with Deep Generative Models
• Transfer Learning for British Sign Language Modelling
• Modeling Taxi Drivers' Behaviour for the Next Destination Prediction
• Quantifying the Tradeoff Between Cybersecurity and Location Privacy
• Hybrid Model with Time Modeling for Sequential Recommender Systems
• A Survey of Techniques All Classifiers Can Learn from Deep Networks: Models, Optimizations, and Regularization
• Application of Bounded Total Variation Denoising in Urban Traffic Analysis
• A Unified Framework for Online Trip Destination Prediction
• Deep Learning based Urban Vehicle Trajectory Analytics