Vector field k-means: Clustering trajectories by fitting multiple vector fields

Nivan Ferreira, James T. Klosowski, Carlos E. Scheidegger, Cláudio T. Silva

Research output: Contribution to journalArticle

Abstract

Scientists study trajectory data to understand trends in movement patterns, such as human mobility for traffic analysis and urban planning. In this paper, we introduce a novel trajectory clustering technique whose central idea is to use vector fields to induce a notion of similarity between trajectories, letting the vector fields themselves define and represent each cluster. We present an efficient algorithm to find a locally optimal clustering of trajectories into vector fields, and demonstrate how vector-field k-means can find patterns missed by previous methods. We present experimental evidence of its effectiveness and efficiency using several datasets, including historical hurricane data, GPS tracks of people and vehicles, and anonymous cellular radio handoffs from a large service provider.

Original languageEnglish (US)
Pages (from-to)201-210
Number of pages10
JournalComputer Graphics Forum
Volume32
Issue number3 PART2
DOIs
StatePublished - Jun 2013

Fingerprint

Trajectories
Urban planning
Hurricanes
Global positioning system

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Vector field k-means : Clustering trajectories by fitting multiple vector fields. / Ferreira, Nivan; Klosowski, James T.; Scheidegger, Carlos E.; Silva, Cláudio T.

In: Computer Graphics Forum, Vol. 32, No. 3 PART2, 06.2013, p. 201-210.

Research output: Contribution to journalArticle

Ferreira, Nivan ; Klosowski, James T. ; Scheidegger, Carlos E. ; Silva, Cláudio T. / Vector field k-means : Clustering trajectories by fitting multiple vector fields. In: Computer Graphics Forum. 2013 ; Vol. 32, No. 3 PART2. pp. 201-210.
@article{67de3d3c4e83448fbb784fcda0a074d7,
title = "Vector field k-means: Clustering trajectories by fitting multiple vector fields",
abstract = "Scientists study trajectory data to understand trends in movement patterns, such as human mobility for traffic analysis and urban planning. In this paper, we introduce a novel trajectory clustering technique whose central idea is to use vector fields to induce a notion of similarity between trajectories, letting the vector fields themselves define and represent each cluster. We present an efficient algorithm to find a locally optimal clustering of trajectories into vector fields, and demonstrate how vector-field k-means can find patterns missed by previous methods. We present experimental evidence of its effectiveness and efficiency using several datasets, including historical hurricane data, GPS tracks of people and vehicles, and anonymous cellular radio handoffs from a large service provider.",
author = "Nivan Ferreira and Klosowski, {James T.} and Scheidegger, {Carlos E.} and Silva, {Cl{\'a}udio T.}",
year = "2013",
month = "6",
doi = "10.1111/cgf.12107",
language = "English (US)",
volume = "32",
pages = "201--210",
journal = "Computer Graphics Forum",
issn = "0167-7055",
publisher = "Wiley-Blackwell",
number = "3 PART2",

}

TY - JOUR

T1 - Vector field k-means

T2 - Clustering trajectories by fitting multiple vector fields

AU - Ferreira, Nivan

AU - Klosowski, James T.

AU - Scheidegger, Carlos E.

AU - Silva, Cláudio T.

PY - 2013/6

Y1 - 2013/6

N2 - Scientists study trajectory data to understand trends in movement patterns, such as human mobility for traffic analysis and urban planning. In this paper, we introduce a novel trajectory clustering technique whose central idea is to use vector fields to induce a notion of similarity between trajectories, letting the vector fields themselves define and represent each cluster. We present an efficient algorithm to find a locally optimal clustering of trajectories into vector fields, and demonstrate how vector-field k-means can find patterns missed by previous methods. We present experimental evidence of its effectiveness and efficiency using several datasets, including historical hurricane data, GPS tracks of people and vehicles, and anonymous cellular radio handoffs from a large service provider.

AB - Scientists study trajectory data to understand trends in movement patterns, such as human mobility for traffic analysis and urban planning. In this paper, we introduce a novel trajectory clustering technique whose central idea is to use vector fields to induce a notion of similarity between trajectories, letting the vector fields themselves define and represent each cluster. We present an efficient algorithm to find a locally optimal clustering of trajectories into vector fields, and demonstrate how vector-field k-means can find patterns missed by previous methods. We present experimental evidence of its effectiveness and efficiency using several datasets, including historical hurricane data, GPS tracks of people and vehicles, and anonymous cellular radio handoffs from a large service provider.

UR - http://www.scopus.com/inward/record.url?scp=84879740999&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879740999&partnerID=8YFLogxK

U2 - 10.1111/cgf.12107

DO - 10.1111/cgf.12107

M3 - Article

VL - 32

SP - 201

EP - 210

JO - Computer Graphics Forum

JF - Computer Graphics Forum

SN - 0167-7055

IS - 3 PART2

ER -