Efficient mining of statistical dependencies

Tim Oates, Matthew D. Schmill, Paul R Cohen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

The Multi-Stream Dependency Detection algorithm finds rules that capture statistical dependencies between patterns in multivariate time series of categorical data [Oates and Cohen, 1996c]. Rule strength is measured by the G statistic [Wickens, 1989], and an upper bound on the value of G for the descendants of a node allows MSDD'S search space to be pruned. However, in the worst case, the algorithm will explore exponentially many rules. This paper presents and empirically evaluates two ways of addressing this problem. The first is a set of three methods for reducing the size of MSDD'S search space based on information collected during the search process. Second, we discuss an implementation of MSDD that distributes its computations over multiple machines on a network.

Original languageEnglish (US)
Title of host publicationIJCAI International Joint Conference on Artificial Intelligence
Pages794-799
Number of pages6
Volume2
StatePublished - 1999
Externally publishedYes
Event16th International Joint Conference on Artificial Intelligence, IJCAI 1999 - Stockholm, Sweden
Duration: Jul 31 1999Aug 6 1999

Other

Other16th International Joint Conference on Artificial Intelligence, IJCAI 1999
CountrySweden
CityStockholm
Period7/31/998/6/99

Fingerprint

Time series
Statistics

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Oates, T., Schmill, M. D., & Cohen, P. R. (1999). Efficient mining of statistical dependencies. In IJCAI International Joint Conference on Artificial Intelligence (Vol. 2, pp. 794-799)

Efficient mining of statistical dependencies. / Oates, Tim; Schmill, Matthew D.; Cohen, Paul R.

IJCAI International Joint Conference on Artificial Intelligence. Vol. 2 1999. p. 794-799.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Oates, T, Schmill, MD & Cohen, PR 1999, Efficient mining of statistical dependencies. in IJCAI International Joint Conference on Artificial Intelligence. vol. 2, pp. 794-799, 16th International Joint Conference on Artificial Intelligence, IJCAI 1999, Stockholm, Sweden, 7/31/99.
Oates T, Schmill MD, Cohen PR. Efficient mining of statistical dependencies. In IJCAI International Joint Conference on Artificial Intelligence. Vol. 2. 1999. p. 794-799
Oates, Tim ; Schmill, Matthew D. ; Cohen, Paul R. / Efficient mining of statistical dependencies. IJCAI International Joint Conference on Artificial Intelligence. Vol. 2 1999. pp. 794-799
@inproceedings{25484608943f4cadb7ed3fb39812e8af,
title = "Efficient mining of statistical dependencies",
abstract = "The Multi-Stream Dependency Detection algorithm finds rules that capture statistical dependencies between patterns in multivariate time series of categorical data [Oates and Cohen, 1996c]. Rule strength is measured by the G statistic [Wickens, 1989], and an upper bound on the value of G for the descendants of a node allows MSDD'S search space to be pruned. However, in the worst case, the algorithm will explore exponentially many rules. This paper presents and empirically evaluates two ways of addressing this problem. The first is a set of three methods for reducing the size of MSDD'S search space based on information collected during the search process. Second, we discuss an implementation of MSDD that distributes its computations over multiple machines on a network.",
author = "Tim Oates and Schmill, {Matthew D.} and Cohen, {Paul R}",
year = "1999",
language = "English (US)",
volume = "2",
pages = "794--799",
booktitle = "IJCAI International Joint Conference on Artificial Intelligence",

}

TY - GEN

T1 - Efficient mining of statistical dependencies

AU - Oates, Tim

AU - Schmill, Matthew D.

AU - Cohen, Paul R

PY - 1999

Y1 - 1999

N2 - The Multi-Stream Dependency Detection algorithm finds rules that capture statistical dependencies between patterns in multivariate time series of categorical data [Oates and Cohen, 1996c]. Rule strength is measured by the G statistic [Wickens, 1989], and an upper bound on the value of G for the descendants of a node allows MSDD'S search space to be pruned. However, in the worst case, the algorithm will explore exponentially many rules. This paper presents and empirically evaluates two ways of addressing this problem. The first is a set of three methods for reducing the size of MSDD'S search space based on information collected during the search process. Second, we discuss an implementation of MSDD that distributes its computations over multiple machines on a network.

AB - The Multi-Stream Dependency Detection algorithm finds rules that capture statistical dependencies between patterns in multivariate time series of categorical data [Oates and Cohen, 1996c]. Rule strength is measured by the G statistic [Wickens, 1989], and an upper bound on the value of G for the descendants of a node allows MSDD'S search space to be pruned. However, in the worst case, the algorithm will explore exponentially many rules. This paper presents and empirically evaluates two ways of addressing this problem. The first is a set of three methods for reducing the size of MSDD'S search space based on information collected during the search process. Second, we discuss an implementation of MSDD that distributes its computations over multiple machines on a network.

UR - http://www.scopus.com/inward/record.url?scp=0001918106&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0001918106&partnerID=8YFLogxK

M3 - Conference contribution

VL - 2

SP - 794

EP - 799

BT - IJCAI International Joint Conference on Artificial Intelligence

ER -