On the detection of concept changes in time-varying data stream by testing exchangeability

Shen Shyang Ho, Harry Wechsler

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

8 Scopus citations

Abstract

A martingale framework for concept change detection based on testing data exchangeability was recently proposed (Ho, 2005). In this paper, we describe the proposed change-detection test based on the Doob's Maximal Inequality and show that it is an approx-imation of the sequential probability ratio test (SPRT). The relationship between the threshold value used in the proposed test and its size and power is deduced from the approximation. The mean delay time before a change is detected is estimated using the average sample number of a SPRT. The performance of the test using various threshold values is examined on five different data stream scenarios simulated using two synthetic data sets. Finally, experimental results show that the test is effective in detecting changes in time-varying data streams simulated using three benchmark data sets.

Original languageEnglish (US)
Title of host publicationProceedings of the 21st Conference on Uncertainty in Artificial Intelligence, UAI 2005
PublisherAUAI Press
Pages267-274
Number of pages8
ISBN (Print)0974903914
StatePublished - 2005
Externally publishedYes
Event21st Conference on Uncertainty in Artificial Intelligence, UAI 2005 - Edinburgh, United Kingdom
Duration: Jul 26 2005Jul 29 2005

Publication series

NameProceedings of the 21st Conference on Uncertainty in Artificial Intelligence, UAI 2005

Conference

Conference21st Conference on Uncertainty in Artificial Intelligence, UAI 2005
Country/TerritoryUnited Kingdom
CityEdinburgh
Period7/26/057/29/05

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'On the detection of concept changes in time-varying data stream by testing exchangeability'. Together they form a unique fingerprint.

Cite this