Long Short-Term Memory Recurrent Neural Network for Urban Traffic Prediction: A Case Study of Seoul

Abstract

Traffic prediction is an important research issue for solving the traffic congestion problems in an Intelligent Transportation System (ITS). In urban areas, traffic congestion has increasingly become a difficult problem. In recent years, abundant traffic data and powerful GPU computing have led to improved accuracy in traffic data analysis via deep learning approaches. In this paper, we propose a long short-term memory recurrent neural network for urban traffic prediction in a case study of Seoul, Korea. The proposed method combines various kinds of time-series data into a model and we conduct comparative analysis using synthetic and real datasets. Our model confirms the proposed method can achieve better accuracy.