Learning From Trajectory Data With MobiML

Anita Graser; Melitta Dragaschnig

Learning From Trajectory Data With MobiML

Anita Graser (Author and Speaker)
, Melitta Dragaschnig

Data Science & Artifical Intelligence

Research output: Chapter in Book or Conference Proceedings › Conference Proceedings with Oral Presentation › peer-review

Abstract

This demo paper introduces MobiML, a new library that aims to help scientists and engineers with developing mobility ML solutions using trajectory data. We also demonstrate how MobiML can speed up ML development workflows using the example of a reproduction of the workflow for training a GeoTrackNet trajectory anomaly detection model. MobiML is available at: https://github.com/movingpandas/mobiml.

Original language	English
Title of host publication	Proceedings of the Workshops of the EDBT/ICDT 2025 Joint Conference
Editors	Matthias Boehm, Khuzaima Daudjee
Number of pages	4
Volume	3946
Publication status	Published - 2025
Event	7th International Workshop on Big Mobility Data Analytics - co-located with the EDBT/ICDT Conference, Barcelona, Spain Duration: 25 Mar 2025 → 28 Mar 2025 https://www.datastories.org/bmda25/index.html

Conference

Conference	7th International Workshop on Big Mobility Data Analytics
Abbreviated title	BMDA2025
Country/Territory	Spain
City	Barcelona
Period	25/03/25 → 28/03/25
Internet address	https://www.datastories.org/bmda25/index.html

Research Field

Multimodal Analytics

Keywords

Machine learning
GeoAI
Mobility Data Science

Access to Document

https://ceur-ws.org/Vol-3946/BMDA-6.pdf

1 Open Source-Software

MobiML: Framework for machine learning from movement data
Graser, A. (Developer), Dragaschnig, M. (Developer) & Agarkova, M. (Developer), 6 Mar 2025
Research output: Software › Open Source-Software

Cite this

@inproceedings{5b08634c07564f579f9d2107b3168c85,

title = "Learning From Trajectory Data With MobiML",

abstract = "This demo paper introduces MobiML, a new library that aims to help scientists and engineers with developing mobility ML solutions using trajectory data. We also demonstrate how MobiML can speed up ML development workflows using the example of a reproduction of the workflow for training a GeoTrackNet trajectory anomaly detection model. MobiML is available at: https://github.com/movingpandas/mobiml.",

keywords = "Machine learning, GeoAI, Mobility Data Science",

author = "Anita Graser and Melitta Dragaschnig",

year = "2025",

language = "English",

volume = "3946",

editor = "Matthias Boehm and Khuzaima Daudjee",

booktitle = "Proceedings of the Workshops of the EDBT/ICDT 2025 Joint Conference",

note = "7th International Workshop on Big Mobility Data Analytics, BMDA2025 ; Conference date: 25-03-2025 Through 28-03-2025",

url = "https://www.datastories.org/bmda25/index.html",

}

TY - GEN

T1 - Learning From Trajectory Data With MobiML

AU - Dragaschnig, Melitta

A2 - Graser, Anita

A2 - Boehm, Matthias

A2 - Daudjee, Khuzaima

PY - 2025

Y1 - 2025

N2 - This demo paper introduces MobiML, a new library that aims to help scientists and engineers with developing mobility ML solutions using trajectory data. We also demonstrate how MobiML can speed up ML development workflows using the example of a reproduction of the workflow for training a GeoTrackNet trajectory anomaly detection model. MobiML is available at: https://github.com/movingpandas/mobiml.

AB - This demo paper introduces MobiML, a new library that aims to help scientists and engineers with developing mobility ML solutions using trajectory data. We also demonstrate how MobiML can speed up ML development workflows using the example of a reproduction of the workflow for training a GeoTrackNet trajectory anomaly detection model. MobiML is available at: https://github.com/movingpandas/mobiml.

KW - Machine learning

KW - GeoAI

KW - Mobility Data Science

M3 - Conference Proceedings with Oral Presentation

VL - 3946

BT - Proceedings of the Workshops of the EDBT/ICDT 2025 Joint Conference

T2 - 7th International Workshop on Big Mobility Data Analytics

Y2 - 25 March 2025 through 28 March 2025

ER -

Learning From Trajectory Data With MobiML

Abstract

Conference

Research Field

Keywords

Access to Document

Fingerprint

Research output

MobiML: Framework for machine learning from movement data

Cite this