Distinguishing Trajectories from Different Drivers using Incompletely Labeled Trajectories

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

4 Citationer (Scopus)

Resumé

We consider a scenario that occurs often in the auto insurance industry. We are given a large collection of trajectories that stem from many different drivers. Only a small number of the trajectories are labeled with driver identifiers, and only some drivers are used in labels. The problem is to label correctly the unlabeled trajectories with driver identifiers. This is important in auto insurance to detect possible fraud and to identify the driver in, e.g., pay-as-you-drive settings when a vehicle has been involved in an incident. To solve the problem, we first propose a Trajectory-to-Image( T2I) encoding scheme that captures both geographic features and driving behavior features of trajectories in 3D images. Next, we propose a multi-task, deep learning model called T2INet for estimating the total number of drivers in the unlabeled trajectories, and then we partition the unlabeled trajectories into groups so that the trajectories in a group belong to the same driver. Experimental results on a large trajectory data set offer insight into the design properties of T2INet and demonstrate that T2INet is capable of outperforming baselines and the state-of-the-art method.
OriginalsprogEngelsk
TitelCIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management
RedaktørerNorman Paton, Selcuk Candan, Haixun Wang, James Allan, Rakesh Agrawal, Alexandros Labrinidis, Alfredo Cuzzocrea, Mohammed Zaki, Divesh Srivastava, Andrei Broder, Assaf Schuster
Antal sider10
ForlagAssociation for Computing Machinery
Publikationsdato17 okt. 2018
Sider863-872
ISBN (Elektronisk)978-1-4503-6014-2
DOI
StatusUdgivet - 17 okt. 2018
Begivenhed27th ACM International Conference on Information and Knowledge Management - Torino, Italien
Varighed: 22 okt. 201826 okt. 2018
http://www.cikm2018.units.it/

Konference

Konference27th ACM International Conference on Information and Knowledge Management
LandItalien
ByTorino
Periode22/10/201826/10/2018
Internetadresse

Citer dette

Kieu, T., Yang, B., Guo, C., & Jensen, C. S. (2018). Distinguishing Trajectories from Different Drivers using Incompletely Labeled Trajectories. I N. Paton, S. Candan, H. Wang, J. Allan, R. Agrawal, A. Labrinidis, A. Cuzzocrea, M. Zaki, D. Srivastava, A. Broder, ... A. Schuster (red.), CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management (s. 863-872). Association for Computing Machinery. https://doi.org/10.1145/3269206.3271762
Kieu, Tung ; Yang, Bin ; Guo, Chenjuan ; Jensen, Christian S. / Distinguishing Trajectories from Different Drivers using Incompletely Labeled Trajectories. CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management. red. / Norman Paton ; Selcuk Candan ; Haixun Wang ; James Allan ; Rakesh Agrawal ; Alexandros Labrinidis ; Alfredo Cuzzocrea ; Mohammed Zaki ; Divesh Srivastava ; Andrei Broder ; Assaf Schuster. Association for Computing Machinery, 2018. s. 863-872
@inproceedings{11c76413eb7841c4bf8de11546cfb2af,
title = "Distinguishing Trajectories from Different Drivers using Incompletely Labeled Trajectories",
abstract = "We consider a scenario that occurs often in the auto insurance industry. We are given a large collection of trajectories that stem from many different drivers. Only a small number of the trajectories are labeled with driver identifiers, and only some drivers are used in labels. The problem is to label correctly the unlabeled trajectories with driver identifiers. This is important in auto insurance to detect possible fraud and to identify the driver in, e.g., pay-as-you-drive settings when a vehicle has been involved in an incident. To solve the problem, we first propose a Trajectory-to-Image( T2I) encoding scheme that captures both geographic features and driving behavior features of trajectories in 3D images. Next, we propose a multi-task, deep learning model called T2INet for estimating the total number of drivers in the unlabeled trajectories, and then we partition the unlabeled trajectories into groups so that the trajectories in a group belong to the same driver. Experimental results on a large trajectory data set offer insight into the design properties of T2INet and demonstrate that T2INet is capable of outperforming baselines and the state-of-the-art method.",
keywords = "Deep learning, Multi-task learning, Representation learning, Trajectory analysis",
author = "Tung Kieu and Bin Yang and Chenjuan Guo and Jensen, {Christian S.}",
year = "2018",
month = "10",
day = "17",
doi = "10.1145/3269206.3271762",
language = "English",
pages = "863--872",
editor = "Norman Paton and Selcuk Candan and Haixun Wang and James Allan and Rakesh Agrawal and Alexandros Labrinidis and Alfredo Cuzzocrea and Mohammed Zaki and Divesh Srivastava and Andrei Broder and Assaf Schuster",
booktitle = "CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management",
publisher = "Association for Computing Machinery",
address = "United States",

}

Kieu, T, Yang, B, Guo, C & Jensen, CS 2018, Distinguishing Trajectories from Different Drivers using Incompletely Labeled Trajectories. i N Paton, S Candan, H Wang, J Allan, R Agrawal, A Labrinidis, A Cuzzocrea, M Zaki, D Srivastava, A Broder & A Schuster (red), CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management. Association for Computing Machinery, s. 863-872, 27th ACM International Conference on Information and Knowledge Management, Torino, Italien, 22/10/2018. https://doi.org/10.1145/3269206.3271762

Distinguishing Trajectories from Different Drivers using Incompletely Labeled Trajectories. / Kieu, Tung; Yang, Bin; Guo, Chenjuan; Jensen, Christian S.

CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management. red. / Norman Paton; Selcuk Candan; Haixun Wang; James Allan; Rakesh Agrawal; Alexandros Labrinidis; Alfredo Cuzzocrea; Mohammed Zaki; Divesh Srivastava; Andrei Broder; Assaf Schuster. Association for Computing Machinery, 2018. s. 863-872.

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

TY - GEN

T1 - Distinguishing Trajectories from Different Drivers using Incompletely Labeled Trajectories

AU - Kieu, Tung

AU - Yang, Bin

AU - Guo, Chenjuan

AU - Jensen, Christian S.

PY - 2018/10/17

Y1 - 2018/10/17

N2 - We consider a scenario that occurs often in the auto insurance industry. We are given a large collection of trajectories that stem from many different drivers. Only a small number of the trajectories are labeled with driver identifiers, and only some drivers are used in labels. The problem is to label correctly the unlabeled trajectories with driver identifiers. This is important in auto insurance to detect possible fraud and to identify the driver in, e.g., pay-as-you-drive settings when a vehicle has been involved in an incident. To solve the problem, we first propose a Trajectory-to-Image( T2I) encoding scheme that captures both geographic features and driving behavior features of trajectories in 3D images. Next, we propose a multi-task, deep learning model called T2INet for estimating the total number of drivers in the unlabeled trajectories, and then we partition the unlabeled trajectories into groups so that the trajectories in a group belong to the same driver. Experimental results on a large trajectory data set offer insight into the design properties of T2INet and demonstrate that T2INet is capable of outperforming baselines and the state-of-the-art method.

AB - We consider a scenario that occurs often in the auto insurance industry. We are given a large collection of trajectories that stem from many different drivers. Only a small number of the trajectories are labeled with driver identifiers, and only some drivers are used in labels. The problem is to label correctly the unlabeled trajectories with driver identifiers. This is important in auto insurance to detect possible fraud and to identify the driver in, e.g., pay-as-you-drive settings when a vehicle has been involved in an incident. To solve the problem, we first propose a Trajectory-to-Image( T2I) encoding scheme that captures both geographic features and driving behavior features of trajectories in 3D images. Next, we propose a multi-task, deep learning model called T2INet for estimating the total number of drivers in the unlabeled trajectories, and then we partition the unlabeled trajectories into groups so that the trajectories in a group belong to the same driver. Experimental results on a large trajectory data set offer insight into the design properties of T2INet and demonstrate that T2INet is capable of outperforming baselines and the state-of-the-art method.

KW - Deep learning

KW - Multi-task learning

KW - Representation learning

KW - Trajectory analysis

UR - http://www.scopus.com/inward/record.url?scp=85058041785&partnerID=8YFLogxK

U2 - 10.1145/3269206.3271762

DO - 10.1145/3269206.3271762

M3 - Article in proceeding

SP - 863

EP - 872

BT - CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management

A2 - Paton, Norman

A2 - Candan, Selcuk

A2 - Wang, Haixun

A2 - Allan, James

A2 - Agrawal, Rakesh

A2 - Labrinidis, Alexandros

A2 - Cuzzocrea, Alfredo

A2 - Zaki, Mohammed

A2 - Srivastava, Divesh

A2 - Broder, Andrei

A2 - Schuster, Assaf

PB - Association for Computing Machinery

ER -

Kieu T, Yang B, Guo C, Jensen CS. Distinguishing Trajectories from Different Drivers using Incompletely Labeled Trajectories. I Paton N, Candan S, Wang H, Allan J, Agrawal R, Labrinidis A, Cuzzocrea A, Zaki M, Srivastava D, Broder A, Schuster A, red., CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management. Association for Computing Machinery. 2018. s. 863-872 https://doi.org/10.1145/3269206.3271762