Research Article
Using Deep Learning to Predict Short Term Traffic Flow: A Systematic Literature Review
@INPROCEEDINGS{10.1007/978-3-319-93710-6_11, author={Usman Ali and Tariq Mahmood}, title={Using Deep Learning to Predict Short Term Traffic Flow: A Systematic Literature Review}, proceedings={Intelligent Transport Systems -- From Research and Development to the Market Uptake. First International Conference, INTSYS 2017, Hyvink\aa{}\aa{}, Finland, November 29-30, 2017, Proceedings}, proceedings_a={INTSYS}, year={2018}, month={7}, keywords={Traffic flow prediction Deep learning Intelligent transport systems Big data}, doi={10.1007/978-3-319-93710-6_11} }- Usman Ali
Tariq Mahmood
Year: 2018
Using Deep Learning to Predict Short Term Traffic Flow: A Systematic Literature Review
INTSYS
Springer
DOI: 10.1007/978-3-319-93710-6_11
Abstract
This paper systematically reviews Deep Learning-based methods for traffic flow prediction. We extracted 26 articles using a concrete methodology and reviewed them from two perspectives: first, the deep learning architecture used; and second, the datasets and data dimensions incorporated. Recent big data explosion caused by sensors, IoV, IoT and GPS technology needs traffic analytics using deep architectures. This survey reveals that the LSTM (Long Short-Term Memory) Neural Networks are the most commonly used architecture for short term traffic flow prediction due to their inherent ability to handle sequential data. Among the datasets, PeMS is the most commonly used for traffic flow prediction task. Today, Intelligent Transport Systems (ITS) are not limited to temporal data; spatial dimension is also incorporated along with weather data, and traffic sentiments from twitter, Facebook and Instagram to get better results. In the authors’ knowledge, this is the first deep learning review in ITS domain.
