Efficient Journey Planning and Congestion Prediction Through Deep Learning

Bin Othman, M. S., Keoh, S. L. and Tan, G. (2017) Efficient Journey Planning and Congestion Prediction Through Deep Learning. In: The Third IEEE Annual International Smart Cities Conference (ISC2 2017), Wuxi, China, 14-17 Sep 2017, ISBN 9781538625248 (doi: 10.1109/ISC2.2017.8090805)

Abstract

The advancements of technology continuously rising over the years has seen many applications that are useful in providing users with sufficient information to make better journey plans on their own. However, commuters still find themselves going through congested routes every day to get to their destinations. This paper attempts to delineate the possibilities of improving urban mobility through big data processing and deep-learning models. Essentially, through a predictive model to predict congestion and its duration, this paper aims to develop and validate a functional journey planning mobile application that can predict traffic conditions, allowing road users to make better informed decisions to their travel plans. This paper proposes a Multi-Layered Perceptron (MLP) deep learning model for congestion prediction and supplements a Linear Regression (LR) model to predict its duration. The proposed MLP-LR model performed reasonably well with an accuracy of 63% in predicting an occurrence of congestion. Some critical discussions on further research opportunities stemming from this study is also presented.

• Item Type: Conference Proceedings
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Keoh, Dr Sye Loong
• Authors: Bin Othman, M. S., Keoh, S. L., and Tan, G.
• College/School: College of Science and Engineering > School of Computing Science
• ISBN: 9781538625248
• Copyright Holders: Copyright © 2017 IEEE
• Publisher Policy: Reproduced in accordance with the copyright policy of the publisher

Related URLs

• Organisation