Adaptable anomaly detection in traffic flow time series

Md Rakibul Alam, Ilias Gerostathopoulos, Sasan Amini, Christian Prehofer, Alessandro Attanasi

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

Abstract

Analysis of traffic data is an essential component of many intelligent transportation system applications where the quality of data plays an important role. Traffic data collected through sensors such as loop detectors often contain anomalies, e.g. due to malfunctioning detectors or anomalous traffic conditions. Regardless of the rooting cause, such data can heavily affect the results of the subsequent analysis (e.g.Traffic prediction). There are several challenges regarding anomaly detection, including absence of universal definition of anomaly, change of traffic pattern over time, as well as unavailability of labeled data, use-case driven analysis. In this paper, a new anomaly detection method for traffic univariate time-series is proposed which does not assume labeled historical data yet uses expert feedback to deal with the fluid definition of anomaly. The method is exemplified and evaluated by applying it on real traffic time series data collected through loop detectors installed in an urban road network in Europe. Employing the proposed method as a pre-process of traffic state estimation can increase the accuracy measure as well as ease the learning of different traffic patterns.

Original languageEnglish
Title of host publicationMT-ITS 2019 - 6th International Conference on Models and Technologies for Intelligent Transportation Systems
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538694848
DOIs
Publication statusPublished - 1 Jun 2019
Externally publishedYes
Event6th International Conference on Models and Technologies for Intelligent Transportation Systems, MT-ITS 2019 - Krakow, Poland
Duration: 5 Jun 20197 Jun 2019

Publication series

NameMT-ITS 2019 - 6th International Conference on Models and Technologies for Intelligent Transportation Systems

Conference

Conference6th International Conference on Models and Technologies for Intelligent Transportation Systems, MT-ITS 2019
CountryPoland
CityKrakow
Period5/06/197/06/19

Keywords

  • anomaly detection
  • clustering
  • loop detectors
  • traffic flow time-series

Fingerprint Dive into the research topics of 'Adaptable anomaly detection in traffic flow time series'. Together they form a unique fingerprint.

Cite this