EPeriodicity: Mining event periodicity from incomplete observations

Zhenhui Li, Jingjing Wang, Jiawei Han

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Advanced technology in GPS and sensors enables us to track physical events, such as human movements and facility usage. Periodicity analysis from the recorded data is an important data mining task which provides useful insights into the physical events and enables us to report outliers and predict future behaviors. To mine periodicity in an event, we have to face real-world challenges of inherently complicated periodic behaviors and imperfect data collection problem. Specifically, the hidden temporal periodic behaviors could be oscillating and noisy, and the observations of the event could be incomplete. In this paper, we propose a novel probabilistic measure for periodicity and design a practical algorithm, ePeriodicity, to detect periods. Our method has thoroughly considered the uncertainties and noises in periodic behaviors and is provably robust to incomplete observations. Comprehensive experiments on both synthetic and real datasets demonstrate the effectiveness of our method.

Original languageEnglish (US)
Article number6940249
Pages (from-to)1219-1232
Number of pages14
JournalIEEE Transactions on Knowledge and Data Engineering
Volume27
Issue number5
DOIs
StatePublished - May 1 2015

Fingerprint

Data mining
Global positioning system
Sensors
Experiments
Uncertainty

All Science Journal Classification (ASJC) codes

  • Information Systems
  • Computer Science Applications
  • Computational Theory and Mathematics

Cite this

@article{3515aa74b558446ea8b06c2fc7ce307e,
title = "EPeriodicity: Mining event periodicity from incomplete observations",
abstract = "Advanced technology in GPS and sensors enables us to track physical events, such as human movements and facility usage. Periodicity analysis from the recorded data is an important data mining task which provides useful insights into the physical events and enables us to report outliers and predict future behaviors. To mine periodicity in an event, we have to face real-world challenges of inherently complicated periodic behaviors and imperfect data collection problem. Specifically, the hidden temporal periodic behaviors could be oscillating and noisy, and the observations of the event could be incomplete. In this paper, we propose a novel probabilistic measure for periodicity and design a practical algorithm, ePeriodicity, to detect periods. Our method has thoroughly considered the uncertainties and noises in periodic behaviors and is provably robust to incomplete observations. Comprehensive experiments on both synthetic and real datasets demonstrate the effectiveness of our method.",
author = "Zhenhui Li and Jingjing Wang and Jiawei Han",
year = "2015",
month = "5",
day = "1",
doi = "10.1109/TKDE.2014.2365801",
language = "English (US)",
volume = "27",
pages = "1219--1232",
journal = "IEEE Transactions on Knowledge and Data Engineering",
issn = "1041-4347",
publisher = "IEEE Computer Society",
number = "5",

}

EPeriodicity : Mining event periodicity from incomplete observations. / Li, Zhenhui; Wang, Jingjing; Han, Jiawei.

In: IEEE Transactions on Knowledge and Data Engineering, Vol. 27, No. 5, 6940249, 01.05.2015, p. 1219-1232.

Research output: Contribution to journalArticle

TY - JOUR

T1 - EPeriodicity

T2 - Mining event periodicity from incomplete observations

AU - Li, Zhenhui

AU - Wang, Jingjing

AU - Han, Jiawei

PY - 2015/5/1

Y1 - 2015/5/1

N2 - Advanced technology in GPS and sensors enables us to track physical events, such as human movements and facility usage. Periodicity analysis from the recorded data is an important data mining task which provides useful insights into the physical events and enables us to report outliers and predict future behaviors. To mine periodicity in an event, we have to face real-world challenges of inherently complicated periodic behaviors and imperfect data collection problem. Specifically, the hidden temporal periodic behaviors could be oscillating and noisy, and the observations of the event could be incomplete. In this paper, we propose a novel probabilistic measure for periodicity and design a practical algorithm, ePeriodicity, to detect periods. Our method has thoroughly considered the uncertainties and noises in periodic behaviors and is provably robust to incomplete observations. Comprehensive experiments on both synthetic and real datasets demonstrate the effectiveness of our method.

AB - Advanced technology in GPS and sensors enables us to track physical events, such as human movements and facility usage. Periodicity analysis from the recorded data is an important data mining task which provides useful insights into the physical events and enables us to report outliers and predict future behaviors. To mine periodicity in an event, we have to face real-world challenges of inherently complicated periodic behaviors and imperfect data collection problem. Specifically, the hidden temporal periodic behaviors could be oscillating and noisy, and the observations of the event could be incomplete. In this paper, we propose a novel probabilistic measure for periodicity and design a practical algorithm, ePeriodicity, to detect periods. Our method has thoroughly considered the uncertainties and noises in periodic behaviors and is provably robust to incomplete observations. Comprehensive experiments on both synthetic and real datasets demonstrate the effectiveness of our method.

UR - http://www.scopus.com/inward/record.url?scp=84926671700&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84926671700&partnerID=8YFLogxK

U2 - 10.1109/TKDE.2014.2365801

DO - 10.1109/TKDE.2014.2365801

M3 - Article

AN - SCOPUS:84926671700

VL - 27

SP - 1219

EP - 1232

JO - IEEE Transactions on Knowledge and Data Engineering

JF - IEEE Transactions on Knowledge and Data Engineering

SN - 1041-4347

IS - 5

M1 - 6940249

ER -