Worst-case analysis of rule discovery

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

2 Citations (Scopus)

Abstract

In this paper, we perform a worst-case analysis of rule discovery. A rule is defined as a probabilistic constraint of true assignment to the class attribute of corresponding examples. In data mining, a rule can be considered as representing an important class of discovered patterns. We accomplish the aforementioned objective by extending a preliminary version of PAC learning, which represents a worst-case analysis for classification. Our analysis consists of two cases: the case in which we try to avoid finding a bad rule, and the case in which we try to avoid overlooking a good rule. Discussions on related works are also provided for PAC learning, multiple comparison, analysis of association rule discovery, and simultaneous reliability evaluation of a discovered rule.

Original languageEnglish
Title of host publicationDiscovery Science - 4th International Conference, DS 2001, Proceedings
EditorsKlaus P. Jantke, Ayumi Shinohara
PublisherSpringer Verlag
Pages365-377
Number of pages13
ISBN (Print)9783540429562
Publication statusPublished - Jan 1 2001
Externally publishedYes
Event4th International Conference on Discovery Science, DS 2001 - Washington, United States
Duration: Nov 25 2001Nov 28 2001

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2226
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other4th International Conference on Discovery Science, DS 2001
CountryUnited States
CityWashington
Period11/25/0111/28/01

Fingerprint

Worst-case Analysis
Association rules
Data mining
PAC Learning
Probabilistic Constraints
Reliability Evaluation
Multiple Comparisons
Association Rules
Data Mining
Assignment
Attribute

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Suzuki, E. (2001). Worst-case analysis of rule discovery. In K. P. Jantke, & A. Shinohara (Eds.), Discovery Science - 4th International Conference, DS 2001, Proceedings (pp. 365-377). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2226). Springer Verlag.

Worst-case analysis of rule discovery. / Suzuki, Einoshin.

Discovery Science - 4th International Conference, DS 2001, Proceedings. ed. / Klaus P. Jantke; Ayumi Shinohara. Springer Verlag, 2001. p. 365-377 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2226).

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

Suzuki, E 2001, Worst-case analysis of rule discovery. in KP Jantke & A Shinohara (eds), Discovery Science - 4th International Conference, DS 2001, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2226, Springer Verlag, pp. 365-377, 4th International Conference on Discovery Science, DS 2001, Washington, United States, 11/25/01.
Suzuki E. Worst-case analysis of rule discovery. In Jantke KP, Shinohara A, editors, Discovery Science - 4th International Conference, DS 2001, Proceedings. Springer Verlag. 2001. p. 365-377. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Suzuki, Einoshin. / Worst-case analysis of rule discovery. Discovery Science - 4th International Conference, DS 2001, Proceedings. editor / Klaus P. Jantke ; Ayumi Shinohara. Springer Verlag, 2001. pp. 365-377 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{4ce6d102ee7c4418ac5c88d4f390efb3,
title = "Worst-case analysis of rule discovery",
abstract = "In this paper, we perform a worst-case analysis of rule discovery. A rule is defined as a probabilistic constraint of true assignment to the class attribute of corresponding examples. In data mining, a rule can be considered as representing an important class of discovered patterns. We accomplish the aforementioned objective by extending a preliminary version of PAC learning, which represents a worst-case analysis for classification. Our analysis consists of two cases: the case in which we try to avoid finding a bad rule, and the case in which we try to avoid overlooking a good rule. Discussions on related works are also provided for PAC learning, multiple comparison, analysis of association rule discovery, and simultaneous reliability evaluation of a discovered rule.",
author = "Einoshin Suzuki",
year = "2001",
month = "1",
day = "1",
language = "English",
isbn = "9783540429562",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "365--377",
editor = "Jantke, {Klaus P.} and Ayumi Shinohara",
booktitle = "Discovery Science - 4th International Conference, DS 2001, Proceedings",
address = "Germany",

}

TY - GEN

T1 - Worst-case analysis of rule discovery

AU - Suzuki, Einoshin

PY - 2001/1/1

Y1 - 2001/1/1

N2 - In this paper, we perform a worst-case analysis of rule discovery. A rule is defined as a probabilistic constraint of true assignment to the class attribute of corresponding examples. In data mining, a rule can be considered as representing an important class of discovered patterns. We accomplish the aforementioned objective by extending a preliminary version of PAC learning, which represents a worst-case analysis for classification. Our analysis consists of two cases: the case in which we try to avoid finding a bad rule, and the case in which we try to avoid overlooking a good rule. Discussions on related works are also provided for PAC learning, multiple comparison, analysis of association rule discovery, and simultaneous reliability evaluation of a discovered rule.

AB - In this paper, we perform a worst-case analysis of rule discovery. A rule is defined as a probabilistic constraint of true assignment to the class attribute of corresponding examples. In data mining, a rule can be considered as representing an important class of discovered patterns. We accomplish the aforementioned objective by extending a preliminary version of PAC learning, which represents a worst-case analysis for classification. Our analysis consists of two cases: the case in which we try to avoid finding a bad rule, and the case in which we try to avoid overlooking a good rule. Discussions on related works are also provided for PAC learning, multiple comparison, analysis of association rule discovery, and simultaneous reliability evaluation of a discovered rule.

UR - http://www.scopus.com/inward/record.url?scp=23044530635&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=23044530635&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:23044530635

SN - 9783540429562

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 365

EP - 377

BT - Discovery Science - 4th International Conference, DS 2001, Proceedings

A2 - Jantke, Klaus P.

A2 - Shinohara, Ayumi

PB - Springer Verlag

ER -