Splitter

Mining finegrained sequential patterns in semantic trajectories

Chao Zhang, Jiawei Han, Lidan Shou, Jiajun Lu, Thomas F. La Porta

Research output: Contribution to journalArticle

70 Citations (Scopus)

Abstract

Driven by the advance of positioning technology and the popularity of location-sharing services, semantic-enriched trajectory data have become unprecedentedly available. The sequential patterns hidden in such data, when properly defined and extracted, can greatly benefit tasks like targeted advertising and urban planning. Unfortunately, classic sequential pattern mining algorithms developed for transactional data cannot effectively mine patterns in semantic trajectories, mainly because the places in the continuous space cannot be regarded as independent "items". Instead, similar places need to be grouped to collaboratively form frequent sequential patterns. That said, it remains a challenging task to mine what we call fine-grained sequential patterns, which must satisfy spatial compactness, semantic consistency and temporal continuity simultaneously. We propose SPLITTER to effectively mine such fine-grained sequential patterns in two steps. In the first step, it retrieves a set of spatially coarse patterns, each attached with a set of trajectory snippets that precisely record the pattern's occurrences in the database. In the second step, SPLITTER breaks each coarse pattern into fine-grained ones in a top-down manner, by progressively detecting dense and compact clusters in a higher-dimensional space spanned by the snippets. SPLITTER uses an effective algorithm called weighted snippet shift to detect such clusters, and leverages a divide-and-conquer strategy to speed up the top-down pattern splitting process. Our experiments on both real and synthetic data sets demonstrate the effectiveness and efficiency of SPLITTER.

Original languageEnglish (US)
Pages (from-to)769-780
Number of pages12
JournalProceedings of the VLDB Endowment
Volume7
Issue number9
DOIs
StatePublished - Jan 1 2014

Fingerprint

Semantics
Trajectories
Urban planning
Marketing
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Science (miscellaneous)
  • Computer Science(all)

Cite this

Zhang, Chao ; Han, Jiawei ; Shou, Lidan ; Lu, Jiajun ; La Porta, Thomas F. / Splitter : Mining finegrained sequential patterns in semantic trajectories. In: Proceedings of the VLDB Endowment. 2014 ; Vol. 7, No. 9. pp. 769-780.
@article{cc098e0f4dbc4f368637e5fddbd3fe8e,
title = "Splitter: Mining finegrained sequential patterns in semantic trajectories",
abstract = "Driven by the advance of positioning technology and the popularity of location-sharing services, semantic-enriched trajectory data have become unprecedentedly available. The sequential patterns hidden in such data, when properly defined and extracted, can greatly benefit tasks like targeted advertising and urban planning. Unfortunately, classic sequential pattern mining algorithms developed for transactional data cannot effectively mine patterns in semantic trajectories, mainly because the places in the continuous space cannot be regarded as independent {"}items{"}. Instead, similar places need to be grouped to collaboratively form frequent sequential patterns. That said, it remains a challenging task to mine what we call fine-grained sequential patterns, which must satisfy spatial compactness, semantic consistency and temporal continuity simultaneously. We propose SPLITTER to effectively mine such fine-grained sequential patterns in two steps. In the first step, it retrieves a set of spatially coarse patterns, each attached with a set of trajectory snippets that precisely record the pattern's occurrences in the database. In the second step, SPLITTER breaks each coarse pattern into fine-grained ones in a top-down manner, by progressively detecting dense and compact clusters in a higher-dimensional space spanned by the snippets. SPLITTER uses an effective algorithm called weighted snippet shift to detect such clusters, and leverages a divide-and-conquer strategy to speed up the top-down pattern splitting process. Our experiments on both real and synthetic data sets demonstrate the effectiveness and efficiency of SPLITTER.",
author = "Chao Zhang and Jiawei Han and Lidan Shou and Jiajun Lu and {La Porta}, {Thomas F.}",
year = "2014",
month = "1",
day = "1",
doi = "10.14778/2732939.2732949",
language = "English (US)",
volume = "7",
pages = "769--780",
journal = "Proceedings of the VLDB Endowment",
issn = "2150-8097",
publisher = "Very Large Data Base Endowment Inc.",
number = "9",

}

Splitter : Mining finegrained sequential patterns in semantic trajectories. / Zhang, Chao; Han, Jiawei; Shou, Lidan; Lu, Jiajun; La Porta, Thomas F.

In: Proceedings of the VLDB Endowment, Vol. 7, No. 9, 01.01.2014, p. 769-780.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Splitter

T2 - Mining finegrained sequential patterns in semantic trajectories

AU - Zhang, Chao

AU - Han, Jiawei

AU - Shou, Lidan

AU - Lu, Jiajun

AU - La Porta, Thomas F.

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Driven by the advance of positioning technology and the popularity of location-sharing services, semantic-enriched trajectory data have become unprecedentedly available. The sequential patterns hidden in such data, when properly defined and extracted, can greatly benefit tasks like targeted advertising and urban planning. Unfortunately, classic sequential pattern mining algorithms developed for transactional data cannot effectively mine patterns in semantic trajectories, mainly because the places in the continuous space cannot be regarded as independent "items". Instead, similar places need to be grouped to collaboratively form frequent sequential patterns. That said, it remains a challenging task to mine what we call fine-grained sequential patterns, which must satisfy spatial compactness, semantic consistency and temporal continuity simultaneously. We propose SPLITTER to effectively mine such fine-grained sequential patterns in two steps. In the first step, it retrieves a set of spatially coarse patterns, each attached with a set of trajectory snippets that precisely record the pattern's occurrences in the database. In the second step, SPLITTER breaks each coarse pattern into fine-grained ones in a top-down manner, by progressively detecting dense and compact clusters in a higher-dimensional space spanned by the snippets. SPLITTER uses an effective algorithm called weighted snippet shift to detect such clusters, and leverages a divide-and-conquer strategy to speed up the top-down pattern splitting process. Our experiments on both real and synthetic data sets demonstrate the effectiveness and efficiency of SPLITTER.

AB - Driven by the advance of positioning technology and the popularity of location-sharing services, semantic-enriched trajectory data have become unprecedentedly available. The sequential patterns hidden in such data, when properly defined and extracted, can greatly benefit tasks like targeted advertising and urban planning. Unfortunately, classic sequential pattern mining algorithms developed for transactional data cannot effectively mine patterns in semantic trajectories, mainly because the places in the continuous space cannot be regarded as independent "items". Instead, similar places need to be grouped to collaboratively form frequent sequential patterns. That said, it remains a challenging task to mine what we call fine-grained sequential patterns, which must satisfy spatial compactness, semantic consistency and temporal continuity simultaneously. We propose SPLITTER to effectively mine such fine-grained sequential patterns in two steps. In the first step, it retrieves a set of spatially coarse patterns, each attached with a set of trajectory snippets that precisely record the pattern's occurrences in the database. In the second step, SPLITTER breaks each coarse pattern into fine-grained ones in a top-down manner, by progressively detecting dense and compact clusters in a higher-dimensional space spanned by the snippets. SPLITTER uses an effective algorithm called weighted snippet shift to detect such clusters, and leverages a divide-and-conquer strategy to speed up the top-down pattern splitting process. Our experiments on both real and synthetic data sets demonstrate the effectiveness and efficiency of SPLITTER.

UR - http://www.scopus.com/inward/record.url?scp=84901768820&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901768820&partnerID=8YFLogxK

U2 - 10.14778/2732939.2732949

DO - 10.14778/2732939.2732949

M3 - Article

VL - 7

SP - 769

EP - 780

JO - Proceedings of the VLDB Endowment

JF - Proceedings of the VLDB Endowment

SN - 2150-8097

IS - 9

ER -