Traffic Accidents Analysis using Self-Organizing Maps and Association Rules for Improved Tourist Safety

Andreas Gregoriades, Andreas Christodoulides

2017

Abstract

Traffic accidents is the most common cause of injury among tourists. This paper presents a method and a tool for analysing historical traffic accident records using data mining techniques for the development of an application that warns tourist drivers of possible accident risks. The knowledge necessary for the specification of the application is based on patterns distilled from spatiotemporal analysis of historical accidents records. Raw accident obtained from Police records, underwent pre-processing and subsequently was integrated with secondary traffic-flow data from a mesoscopic simulation. Two data mining techniques were applied on the resulting dataset, namely, clustering with self-organizing maps (SOM) and association rules. The former was used to identify accident black spots, while the latter was applied in the clusters that emerged from SOM to identify causes of accidents in each black spot. Identified patterns were utilized to develop a software application to alert travellers of imminent accident risks, using characteristics of drivers along with real-time feeds of drivers’ geolocation and environmental conditions.

Paper Citation

in Harvard Style

Gregoriades A. and Christodoulides A. (2017). Traffic Accidents Analysis using Self-Organizing Maps and Association Rules for Improved Tourist Safety . In Proceedings of the 19th International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-758-247-9, pages 452-459. DOI: 10.5220/0006356204520459

in Bibtex Style

@conference{iceis17,
author={Andreas Gregoriades and Andreas Christodoulides},
title={Traffic Accidents Analysis using Self-Organizing Maps and Association Rules for Improved Tourist Safety},
booktitle={Proceedings of the 19th International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2017},
pages={452-459},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006356204520459},
isbn={978-989-758-247-9},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 19th International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - Traffic Accidents Analysis using Self-Organizing Maps and Association Rules for Improved Tourist Safety
SN - 978-989-758-247-9
AU - Gregoriades A.
AU - Christodoulides A.
PY - 2017
SP - 452
EP - 459
DO - 10.5220/0006356204520459