Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining

Florian Barth*, Stefan Funke*, Tobias Skovgaard Jepsen*, Claudius Proissl*

*Kontaktforfatter

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

2 Downloads (Pure)

Abstrakt

We present analysis techniques for large trajectory data sets that aim to provide a semantic understanding of trajectories reaching beyond them being point sequences in time and space. The presented techniques use a driving preference model w.r.t. road segment traversal costs, e.g., travel time and distance, to analyze and explain trajectories.

In particular, we present trajectory mining techniques that can (a) find interesting points within a trajectory indicating, e.g., a viapoint, and (b) recover the driving preferences of a driver based on their chosen trajectory. We evaluate our techniques on the tasks of viapoint identification and personalized routing using a data set of more than 1 million vehicle trajectories collected throughout Denmark during a 3-year period. Our techniques can be implemented efficiently and are highly parallelizable, allowing them to scale to millions or billions of trajectories.
OriginalsprogEngelsk
TitelBIGSPATIAL '20: Proceedings of the 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data
RedaktørerVarun Chandola, Ranga Raju Vatsavai, Ashwin Shashidharan
Antal sider10
ForlagAssociation for Computing Machinery
Publikationsdato3 nov. 2020
Sider1-10
Artikelnummer6
ISBN (Elektronisk)9781450381628
DOI
StatusUdgivet - 3 nov. 2020
BegivenhedThe 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data -
Varighed: 3 nov. 20203 nov. 2020
https://bigspatial2020.github.io/

Konference

KonferenceThe 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data
Periode03/11/202003/11/2020
Internetadresse

Fingeraftryk Dyk ned i forskningsemnerne om 'Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining'. Sammen danner de et unikt fingeraftryk.

Citationsformater