Incremental, online, and merge mining of partial periodic patterns in time-series databases

Walid G. Aref, Mohamed G. Elfeky, Ahmed Elmagarmid

Research output: Contribution to journalArticle

62 Citations (Scopus)

Abstract

Mining of periodic patterns in time-series databases is an interesting data mining problem, it can be envisioned as a tool for forecasting and prediction of the future behavior of time-series data. Incremental mining refers to the issue of maintaining the discovered patterns over time in the presence of more items being added into the database. Because of the mostly append only nature of updating time-series data, incremental mining would be very effective and efficient. Several algorithms for incremental mining of partial periodic patterns in time-series databases are proposed and are analyzed empirically. The new algorithms allow for online adaptation of the thresholds in order to produce interactive mining of partial periodic patterns. The storage overhead of the incremental online mining algorithms is analyzed. Results show that the storage overhead for storing the intermediate data structures pays off as the incremental online mining of partial periodic patterns proves to be significantly more efficient than the nonincremental nononline versions. Moreover, a new problem, termed merge mining, is introduced as a generalization of incremental mining. Merge mining can be defined as merging the discovered patterns of two or more databases that are mined independently of each other. An algorithm for merge mining of partial periodic patterns in time-series databases is proposed and analyzed.

Original languageEnglish
Pages (from-to)332-342
Number of pages11
JournalIEEE Transactions on Knowledge and Data Engineering
Volume16
Issue number3
DOIs
Publication statusPublished - 1 Mar 2004
Externally publishedYes

Fingerprint

Time series
Data mining
Merging
Data structures

Keywords

  • Data mining
  • Incremental mining
  • Online mining
  • Time-series databases

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Electrical and Electronic Engineering
  • Artificial Intelligence
  • Information Systems

Cite this

Incremental, online, and merge mining of partial periodic patterns in time-series databases. / Aref, Walid G.; Elfeky, Mohamed G.; Elmagarmid, Ahmed.

In: IEEE Transactions on Knowledge and Data Engineering, Vol. 16, No. 3, 01.03.2004, p. 332-342.

Research output: Contribution to journalArticle

@article{1fa2129880074824bb89339f43dfba9f,
title = "Incremental, online, and merge mining of partial periodic patterns in time-series databases",
abstract = "Mining of periodic patterns in time-series databases is an interesting data mining problem, it can be envisioned as a tool for forecasting and prediction of the future behavior of time-series data. Incremental mining refers to the issue of maintaining the discovered patterns over time in the presence of more items being added into the database. Because of the mostly append only nature of updating time-series data, incremental mining would be very effective and efficient. Several algorithms for incremental mining of partial periodic patterns in time-series databases are proposed and are analyzed empirically. The new algorithms allow for online adaptation of the thresholds in order to produce interactive mining of partial periodic patterns. The storage overhead of the incremental online mining algorithms is analyzed. Results show that the storage overhead for storing the intermediate data structures pays off as the incremental online mining of partial periodic patterns proves to be significantly more efficient than the nonincremental nononline versions. Moreover, a new problem, termed merge mining, is introduced as a generalization of incremental mining. Merge mining can be defined as merging the discovered patterns of two or more databases that are mined independently of each other. An algorithm for merge mining of partial periodic patterns in time-series databases is proposed and analyzed.",
keywords = "Data mining, Incremental mining, Online mining, Time-series databases",
author = "Aref, {Walid G.} and Elfeky, {Mohamed G.} and Ahmed Elmagarmid",
year = "2004",
month = "3",
day = "1",
doi = "10.1109/TKDE.2003.1262186",
language = "English",
volume = "16",
pages = "332--342",
journal = "IEEE Transactions on Knowledge and Data Engineering",
issn = "1041-4347",
publisher = "IEEE Computer Society",
number = "3",

}

TY - JOUR

T1 - Incremental, online, and merge mining of partial periodic patterns in time-series databases

AU - Aref, Walid G.

AU - Elfeky, Mohamed G.

AU - Elmagarmid, Ahmed

PY - 2004/3/1

Y1 - 2004/3/1

N2 - Mining of periodic patterns in time-series databases is an interesting data mining problem, it can be envisioned as a tool for forecasting and prediction of the future behavior of time-series data. Incremental mining refers to the issue of maintaining the discovered patterns over time in the presence of more items being added into the database. Because of the mostly append only nature of updating time-series data, incremental mining would be very effective and efficient. Several algorithms for incremental mining of partial periodic patterns in time-series databases are proposed and are analyzed empirically. The new algorithms allow for online adaptation of the thresholds in order to produce interactive mining of partial periodic patterns. The storage overhead of the incremental online mining algorithms is analyzed. Results show that the storage overhead for storing the intermediate data structures pays off as the incremental online mining of partial periodic patterns proves to be significantly more efficient than the nonincremental nononline versions. Moreover, a new problem, termed merge mining, is introduced as a generalization of incremental mining. Merge mining can be defined as merging the discovered patterns of two or more databases that are mined independently of each other. An algorithm for merge mining of partial periodic patterns in time-series databases is proposed and analyzed.

AB - Mining of periodic patterns in time-series databases is an interesting data mining problem, it can be envisioned as a tool for forecasting and prediction of the future behavior of time-series data. Incremental mining refers to the issue of maintaining the discovered patterns over time in the presence of more items being added into the database. Because of the mostly append only nature of updating time-series data, incremental mining would be very effective and efficient. Several algorithms for incremental mining of partial periodic patterns in time-series databases are proposed and are analyzed empirically. The new algorithms allow for online adaptation of the thresholds in order to produce interactive mining of partial periodic patterns. The storage overhead of the incremental online mining algorithms is analyzed. Results show that the storage overhead for storing the intermediate data structures pays off as the incremental online mining of partial periodic patterns proves to be significantly more efficient than the nonincremental nononline versions. Moreover, a new problem, termed merge mining, is introduced as a generalization of incremental mining. Merge mining can be defined as merging the discovered patterns of two or more databases that are mined independently of each other. An algorithm for merge mining of partial periodic patterns in time-series databases is proposed and analyzed.

KW - Data mining

KW - Incremental mining

KW - Online mining

KW - Time-series databases

UR - http://www.scopus.com/inward/record.url?scp=1842530903&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1842530903&partnerID=8YFLogxK

U2 - 10.1109/TKDE.2003.1262186

DO - 10.1109/TKDE.2003.1262186

M3 - Article

VL - 16

SP - 332

EP - 342

JO - IEEE Transactions on Knowledge and Data Engineering

JF - IEEE Transactions on Knowledge and Data Engineering

SN - 1041-4347

IS - 3

ER -