Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining

Florian Barth; Stefan Funke; Tobias Skovgaard Jepsen; Claudius Proissl

doi:10.1145/3423336.3429348

Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining

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

^*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

4 Citations (Scopus)

30 Downloads (Pure)

Abstract

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.

Original language	English
Title of host publication	BIGSPATIAL '20 : Proceedings of the 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data
Editors	Varun Chandola, Ranga Raju Vatsavai, Ashwin Shashidharan
Number of pages	10
Publisher	Association for Computing Machinery
Publication date	3 Nov 2020
Pages	1-10
Article number	6
ISBN (Electronic)	9781450381628
DOIs	https://doi.org/10.1145/3423336.3429348
Publication status	Published - 3 Nov 2020
Event	The 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data - Duration: 3 Nov 2020 → 3 Nov 2020 https://bigspatial2020.github.io/

Conference

Conference	The 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data
Period	03/11/2020 → 03/11/2020
Internet address	https://bigspatial2020.github.io/

Keywords

Personalized Routing
Driving Preference Mining
Transportation
Trajectory Segmentation

Access to Document

10.1145/3423336.3429348

TrajectoryAnalysisSubmitted manuscript, 1.41 MB

https://arxiv.org/abs/2011.03331

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

Barth, F., Funke, S., Skovgaard Jepsen, T., & Proissl, C. (2020). Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining. In V. Chandola, R. R. Vatsavai, & A. Shashidharan (Eds.), BIGSPATIAL '20: Proceedings of the 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data (pp. 1-10). Article 6 Association for Computing Machinery. https://doi.org/10.1145/3423336.3429348

Barth, Florian ; Funke, Stefan ; Skovgaard Jepsen, Tobias et al. / Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining. BIGSPATIAL '20: Proceedings of the 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data. editor / Varun Chandola ; Ranga Raju Vatsavai ; Ashwin Shashidharan. Association for Computing Machinery, 2020. pp. 1-10

@inproceedings{479e2e22ae3e4ed480e8f62c757b23ef,

title = "Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining",

abstract = "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.",

keywords = "Personalized Routing, Driving Preference Mining, Transportation, Trajectory Segmentation",

author = "Florian Barth and Stefan Funke and {Skovgaard Jepsen}, Tobias and Claudius Proissl",

year = "2020",

month = nov,

day = "3",

doi = "10.1145/3423336.3429348",

language = "English",

pages = "1--10",

editor = "Varun Chandola and Vatsavai, {Ranga Raju} and Ashwin Shashidharan",

booktitle = "BIGSPATIAL '20",

publisher = "Association for Computing Machinery",

address = "United States",

note = "The 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data, BigSpatial 2020 ; Conference date: 03-11-2020 Through 03-11-2020",

url = "https://bigspatial2020.github.io/",

}

Barth, F, Funke, S, Skovgaard Jepsen, T & Proissl, C 2020, Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining. in V Chandola, RR Vatsavai & A Shashidharan (eds), BIGSPATIAL '20: Proceedings of the 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data., 6, Association for Computing Machinery, pp. 1-10, The 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data, 03/11/2020. https://doi.org/10.1145/3423336.3429348

Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining. / Barth, Florian; Funke, Stefan; Skovgaard Jepsen, Tobias et al.
BIGSPATIAL '20: Proceedings of the 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data. ed. / Varun Chandola; Ranga Raju Vatsavai; Ashwin Shashidharan. Association for Computing Machinery, 2020. p. 1-10 6.

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining

AU - Barth, Florian

AU - Funke, Stefan

AU - Skovgaard Jepsen, Tobias

AU - Proissl, Claudius

PY - 2020/11/3

Y1 - 2020/11/3

N2 - 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.

AB - 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.

KW - Personalized Routing

KW - Driving Preference Mining

KW - Transportation

KW - Trajectory Segmentation

UR - http://www.scopus.com/inward/record.url?scp=85097575597&partnerID=8YFLogxK

U2 - 10.1145/3423336.3429348

DO - 10.1145/3423336.3429348

M3 - Article in proceeding

SP - 1

EP - 10

BT - BIGSPATIAL '20

A2 - Chandola, Varun

A2 - Vatsavai, Ranga Raju

A2 - Shashidharan, Ashwin

PB - Association for Computing Machinery

T2 - The 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data

Y2 - 3 November 2020 through 3 November 2020

ER -

Barth F, Funke S, Skovgaard Jepsen T, Proissl C. Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining. In Chandola V, Vatsavai RR, Shashidharan A, editors, BIGSPATIAL '20: Proceedings of the 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data. Association for Computing Machinery. 2020. p. 1-10. 6 doi: 10.1145/3423336.3429348

Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining

Abstract

Conference

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Relational Fusion Networks: Graph Convolutional Networks for Road Networks

Towards Data-Efficient Mobility Analytics in Spatial Networks

Graph Convolutional Networks for Road Networks

Cite this