Undirected exception rule discovery as local pattern detection

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

Abstract

In this paper, we give an interpretation of our undirected exception rule discovery as local pattern detection and introduce some of our endeavors. Our undirected exception rule discovery outputs a set of rule pairs, each of which represents a pair of strong rule and its exception rule. A local pattern is defined as a pattern which deviates from a global model, and can be considered to correspond to our exception rule if the global model corresponds to our strong rule. Several attempts for undirected exception rule discovery are introduced in the context of local pattern detection. Our results mainly concern interestingness measure, algorithmic issues, noise modeling, and performance evaluation.

Original languageEnglish
Title of host publicationLocal Pattern Detection - International Seminar, Revised Selected Papers
PublisherSpringer Verlag
Pages207-216
Number of pages10
ISBN (Print)3540265430, 9783540265436
DOIs
Publication statusPublished - Jan 1 2005
Externally publishedYes
EventInternational Seminar on Local Pattern Detection - Dagstuhl, Castle, Germany
Duration: Apr 12 2004Apr 16 2004

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3539 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherInternational Seminar on Local Pattern Detection
CountryGermany
CityDagstuhl, Castle
Period4/12/044/16/04

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Undirected exception rule discovery as local pattern detection'. Together they form a unique fingerprint.

Cite this