A conflict-based confidence measure for associative classification

Peerapon Vateekul, Mei-Ling Shyu

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

3 Citations (Scopus)

Abstract

Associative classification has aroused significant attention recently and achieved promising results. In the rule ranking process, the confidence measure is usually used to sort the class association rules (CARs). However, it may be not good enough for a classification task due to a low discrimination power to instances in the other classes. In this paper, we propose a novel conflict-based confidence measure with an interleaving ranking strategy for re-ranking CARs in an associative classification framework, which better captures the conflict between a rule and a training data instance. In the experiments, the traditional confidence measure and our proposed conflict-based confidence measure with the interleaving ranking strategy are applied as the primary sorting criterion for CARs. The experimental results show that the proposed associative classification framework achieves promising classification accuracy with the use of the conflict-based confidence measure, particularly for an imbalanced data set.

Original languageEnglish
Title of host publication2008 IEEE International Conference on Information Reuse and Integration, IEEE IRI-2008
Pages256-261
Number of pages6
DOIs
StatePublished - Sep 23 2008
Event2008 IEEE International Conference on Information Reuse and Integration, IEEE IRI-2008 - Las Vegas, NV, United States
Duration: Jul 13 2008Jul 15 2008

Other

Other2008 IEEE International Conference on Information Reuse and Integration, IEEE IRI-2008
CountryUnited States
CityLas Vegas, NV
Period7/13/087/15/08

Fingerprint

Association rules
Sorting
Experiments

ASJC Scopus subject areas

  • Information Systems
  • Software

Cite this

Vateekul, P., & Shyu, M-L. (2008). A conflict-based confidence measure for associative classification. In 2008 IEEE International Conference on Information Reuse and Integration, IEEE IRI-2008 (pp. 256-261). [4583039] https://doi.org/10.1109/IRI.2008.4583039

A conflict-based confidence measure for associative classification. / Vateekul, Peerapon; Shyu, Mei-Ling.

2008 IEEE International Conference on Information Reuse and Integration, IEEE IRI-2008. 2008. p. 256-261 4583039.

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

Vateekul, P & Shyu, M-L 2008, A conflict-based confidence measure for associative classification. in 2008 IEEE International Conference on Information Reuse and Integration, IEEE IRI-2008., 4583039, pp. 256-261, 2008 IEEE International Conference on Information Reuse and Integration, IEEE IRI-2008, Las Vegas, NV, United States, 7/13/08. https://doi.org/10.1109/IRI.2008.4583039
Vateekul P, Shyu M-L. A conflict-based confidence measure for associative classification. In 2008 IEEE International Conference on Information Reuse and Integration, IEEE IRI-2008. 2008. p. 256-261. 4583039 https://doi.org/10.1109/IRI.2008.4583039
Vateekul, Peerapon ; Shyu, Mei-Ling. / A conflict-based confidence measure for associative classification. 2008 IEEE International Conference on Information Reuse and Integration, IEEE IRI-2008. 2008. pp. 256-261
@inproceedings{d7467f69ce3d4144862cfa3c1dbf3f85,
title = "A conflict-based confidence measure for associative classification",
abstract = "Associative classification has aroused significant attention recently and achieved promising results. In the rule ranking process, the confidence measure is usually used to sort the class association rules (CARs). However, it may be not good enough for a classification task due to a low discrimination power to instances in the other classes. In this paper, we propose a novel conflict-based confidence measure with an interleaving ranking strategy for re-ranking CARs in an associative classification framework, which better captures the conflict between a rule and a training data instance. In the experiments, the traditional confidence measure and our proposed conflict-based confidence measure with the interleaving ranking strategy are applied as the primary sorting criterion for CARs. The experimental results show that the proposed associative classification framework achieves promising classification accuracy with the use of the conflict-based confidence measure, particularly for an imbalanced data set.",
author = "Peerapon Vateekul and Mei-Ling Shyu",
year = "2008",
month = "9",
day = "23",
doi = "10.1109/IRI.2008.4583039",
language = "English",
isbn = "9781424426607",
pages = "256--261",
booktitle = "2008 IEEE International Conference on Information Reuse and Integration, IEEE IRI-2008",

}

TY - GEN

T1 - A conflict-based confidence measure for associative classification

AU - Vateekul, Peerapon

AU - Shyu, Mei-Ling

PY - 2008/9/23

Y1 - 2008/9/23

N2 - Associative classification has aroused significant attention recently and achieved promising results. In the rule ranking process, the confidence measure is usually used to sort the class association rules (CARs). However, it may be not good enough for a classification task due to a low discrimination power to instances in the other classes. In this paper, we propose a novel conflict-based confidence measure with an interleaving ranking strategy for re-ranking CARs in an associative classification framework, which better captures the conflict between a rule and a training data instance. In the experiments, the traditional confidence measure and our proposed conflict-based confidence measure with the interleaving ranking strategy are applied as the primary sorting criterion for CARs. The experimental results show that the proposed associative classification framework achieves promising classification accuracy with the use of the conflict-based confidence measure, particularly for an imbalanced data set.

AB - Associative classification has aroused significant attention recently and achieved promising results. In the rule ranking process, the confidence measure is usually used to sort the class association rules (CARs). However, it may be not good enough for a classification task due to a low discrimination power to instances in the other classes. In this paper, we propose a novel conflict-based confidence measure with an interleaving ranking strategy for re-ranking CARs in an associative classification framework, which better captures the conflict between a rule and a training data instance. In the experiments, the traditional confidence measure and our proposed conflict-based confidence measure with the interleaving ranking strategy are applied as the primary sorting criterion for CARs. The experimental results show that the proposed associative classification framework achieves promising classification accuracy with the use of the conflict-based confidence measure, particularly for an imbalanced data set.

UR - http://www.scopus.com/inward/record.url?scp=51949096517&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51949096517&partnerID=8YFLogxK

U2 - 10.1109/IRI.2008.4583039

DO - 10.1109/IRI.2008.4583039

M3 - Conference contribution

SN - 9781424426607

SP - 256

EP - 261

BT - 2008 IEEE International Conference on Information Reuse and Integration, IEEE IRI-2008

ER -