Deep Learning from Trajectory Data: a Review of Deep Neural Networks and the Trajectory Data Representations to Train Them

Anita Graser; Anahid Naghibzadeh-Jalali; Jasmin Lampert; Axel Weißenfeld; Krzysztof Janowicz

Deep Learning from Trajectory Data: a Review of Deep Neural Networks and the Trajectory Data Representations to Train Them

Anita Graser (Speaker), Anahid Naghibzadeh-Jalali, Jasmin Lampert, Axel Weißenfeld, Krzysztof Janowicz

Data Science & Artifical Intelligence

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

Abstract

Trajectory data combines the complexities of time series, spatial data, and (sometimes irrational) movement behavior. As data
availability and computing power have increased, so has the popularity of deep learning from trajectory data. This paper aims to provide an overview of deep neural networks designed to learn from trajectory data, focusing on recent work published between 2020 and 2022. We take a data-centric approach and distinguish between deep learning models trained using dense trajectories (quasi-continuous tracking data), sparse trajectories (such as check-in data), and aggregated trajectories (crowd
information).

Original language	English
Title of host publication	CEUR Workshop Proceedings 2023
Editors	George Fletcher, Verena Kantere
Number of pages	12
Volume	3379
Publication status	Published - 2023
Event	Big Mobility Data Analytics BMDA 2023 - Ioannina, Greece Duration: 28 Mar 2023 → 28 Mar 2023

Conference

Conference	Big Mobility Data Analytics BMDA 2023
Country/Territory	Greece
City	Ioannina
Period	28/03/23 → 28/03/23

Research Field

Former Research Field - Data Science

Access to Document

https://ceur-ws.org/Vol-3379/BMDA_2023_paper_7556.pdf

Cite this

@inproceedings{16a6cd61705748f9a662275abf5c9406,

title = "Deep Learning from Trajectory Data: a Review of Deep Neural Networks and the Trajectory Data Representations to Train Them",

abstract = "Trajectory data combines the complexities of time series, spatial data, and (sometimes irrational) movement behavior. As dataavailability and computing power have increased, so has the popularity of deep learning from trajectory data. This paper aims to provide an overview of deep neural networks designed to learn from trajectory data, focusing on recent work published between 2020 and 2022. We take a data-centric approach and distinguish between deep learning models trained using dense trajectories (quasi-continuous tracking data), sparse trajectories (such as check-in data), and aggregated trajectories (crowdinformation).",

author = "Anita Graser and Anahid Naghibzadeh-Jalali and Jasmin Lampert and Axel Wei{\ss}enfeld and Krzysztof Janowicz",

year = "2023",

language = "English",

volume = "3379",

editor = "Fletcher, {George } and Kantere, {Verena }",

booktitle = "CEUR Workshop Proceedings 2023",

note = "Big Mobility Data Analytics BMDA 2023 ; Conference date: 28-03-2023 Through 28-03-2023",

}

Graser, A , Naghibzadeh-Jalali, A , Lampert, J , Weißenfeld, A & Janowicz, K 2023, Deep Learning from Trajectory Data: a Review of Deep Neural Networks and the Trajectory Data Representations to Train Them. in G Fletcher & V Kantere (eds), CEUR Workshop Proceedings 2023. vol. 3379, Big Mobility Data Analytics BMDA 2023, Ioannina, Greece, 28/03/23. <https://ceur-ws.org/Vol-3379/BMDA_2023_paper_7556.pdf>

Deep Learning from Trajectory Data: a Review of Deep Neural Networks and the Trajectory Data Representations to Train Them. / Graser, Anita (Speaker); Naghibzadeh-Jalali, Anahid ; Lampert, Jasmin et al.
CEUR Workshop Proceedings 2023. ed. / George Fletcher; Verena Kantere. Vol. 3379 2023.

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

TY - GEN

T1 - Deep Learning from Trajectory Data

T2 - Big Mobility Data Analytics BMDA 2023

AU - Naghibzadeh-Jalali, Anahid

AU - Lampert, Jasmin

AU - Weißenfeld, Axel

AU - Janowicz, Krzysztof

A2 - Graser, Anita

A2 - Fletcher, George

A2 - Kantere, Verena

PY - 2023

Y1 - 2023

N2 - Trajectory data combines the complexities of time series, spatial data, and (sometimes irrational) movement behavior. As dataavailability and computing power have increased, so has the popularity of deep learning from trajectory data. This paper aims to provide an overview of deep neural networks designed to learn from trajectory data, focusing on recent work published between 2020 and 2022. We take a data-centric approach and distinguish between deep learning models trained using dense trajectories (quasi-continuous tracking data), sparse trajectories (such as check-in data), and aggregated trajectories (crowdinformation).

AB - Trajectory data combines the complexities of time series, spatial data, and (sometimes irrational) movement behavior. As dataavailability and computing power have increased, so has the popularity of deep learning from trajectory data. This paper aims to provide an overview of deep neural networks designed to learn from trajectory data, focusing on recent work published between 2020 and 2022. We take a data-centric approach and distinguish between deep learning models trained using dense trajectories (quasi-continuous tracking data), sparse trajectories (such as check-in data), and aggregated trajectories (crowdinformation).

M3 - Conference Proceedings with Oral Presentation

VL - 3379

BT - CEUR Workshop Proceedings 2023

Y2 - 28 March 2023 through 28 March 2023

ER -

Deep Learning from Trajectory Data: a Review of Deep Neural Networks and the Trajectory Data Representations to Train Them

Abstract

Conference

Research Field

Access to Document

Fingerprint

Cite this