Mining positional data streams

Jens Haase; Ulf Brefeld

doi:10.1007/978-3-319-17876-9_7

Mining positional data streams

Jens Haase
, Ulf Brefeld^*

^*Korrespondierende/r Autor/-in für diese Arbeit

Technische Universität Darmstadt

Publikation: Beiträge in Sammelwerken › Aufsätze in Konferenzbänden › Forschung › Begutachtung

7 Zitate (Scopus)

Abstract

We study frequent pattern mining from positional data streams. Existing approaches require discretised data to identify atomic events and are not applicable in our continuous setting. We propose an efficient trajectory-based preprocessing to identify similar movements and a distributed pattern mining algorithm to identify frequent trajectories. We empirically evaluate all parts of the processing pipeline.

Originalsprache	Englisch
Titel	New Frontiers in Mining Complex Patterns
Seitenumfang	15
Herausgeber (Verlag)	Springer Verlag
Erscheinungsdatum	2015
Seiten	102-116
ISBN (Print)	978-3-319-17875-2
ISBN (elektronisch)	978-3-319-17876-9
DOIs	https://doi.org/10.1007/978-3-319-17876-9_7
Publikationsstatus	Erschienen - 2015
Extern publiziert	Ja
Veranstaltung	3rd International Workshop on New Frontiers in Mining Complex Patterns - NFMCP 2014 - Nancy, Frankreich Dauer: 19.09.2014 → 19.09.2014 Konferenznummer: 3

Fachgebiete und Schlagwörter

Informatik
Wirtschaftsinformatik

Zugriff auf Dokument

10.1007/978-3-319-17876-9_7

Dieses zitieren

@inbook{a4f6db1289c04684bd8c28473bb637bf,

title = "Mining positional data streams",

abstract = "We study frequent pattern mining from positional data streams. Existing approaches require discretised data to identify atomic events and are not applicable in our continuous setting. We propose an efficient trajectory-based preprocessing to identify similar movements and a distributed pattern mining algorithm to identify frequent trajectories. We empirically evaluate all parts of the processing pipeline.",

keywords = "Informatics, Pattern Mining, Dynamic Time Warping, Positional Data, Frequent Episode, Event Stream, Business informatics",

author = "Jens Haase and Ulf Brefeld",

year = "2015",

doi = "10.1007/978-3-319-17876-9\_7",

language = "English",

isbn = "978-3-319-17875-2",

pages = "102--116",

booktitle = "New Frontiers in Mining Complex Patterns",

publisher = "Springer Verlag",

address = "Germany",

note = "3rd International Workshop on New Frontiers in Mining Complex Patterns - NFMCP 2014, NFMCP 2014 ; Conference date: 19-09-2014 Through 19-09-2014",

}

TY - CHAP

T1 - Mining positional data streams

AU - Haase, Jens

AU - Brefeld, Ulf

N1 - Conference code: 3

PY - 2015

Y1 - 2015

N2 - We study frequent pattern mining from positional data streams. Existing approaches require discretised data to identify atomic events and are not applicable in our continuous setting. We propose an efficient trajectory-based preprocessing to identify similar movements and a distributed pattern mining algorithm to identify frequent trajectories. We empirically evaluate all parts of the processing pipeline.

AB - We study frequent pattern mining from positional data streams. Existing approaches require discretised data to identify atomic events and are not applicable in our continuous setting. We propose an efficient trajectory-based preprocessing to identify similar movements and a distributed pattern mining algorithm to identify frequent trajectories. We empirically evaluate all parts of the processing pipeline.

KW - Informatics

KW - Pattern Mining

KW - Dynamic Time Warping

KW - Positional Data

KW - Frequent Episode

KW - Event Stream

KW - Business informatics

UR - https://www.scopus.com/pages/publications/84931056379

U2 - 10.1007/978-3-319-17876-9_7

DO - 10.1007/978-3-319-17876-9_7

M3 - Article in conference proceedings

AN - SCOPUS:84931056379

SN - 978-3-319-17875-2

SP - 102

EP - 116

BT - New Frontiers in Mining Complex Patterns

PB - Springer Verlag

T2 - 3rd International Workshop on New Frontiers in Mining Complex Patterns - NFMCP 2014

Y2 - 19 September 2014 through 19 September 2014

ER -

Mining positional data streams

Abstract

Fachgebiete und Schlagwörter

Zugriff auf Dokument

Fingerprint

Dieses zitieren