CCCS: A top-down associative classifier for imbalanced class distribution

Bavani Arunasalam; Sanjay Chawla

CCCS: A top-down associative classifier for imbalanced class distribution

Bavani Arunasalam, Sanjay Chawla

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

45 Scopus citations

Abstract

In this paper we propose CCCS, a new algorithm for classification based on association rule mining. The key innovation in CCCS is the use of a new measure, the "Complement Class Support (CCS)" whose application results in rules which are guaranteed to be positively correlated. Furthermore, the anti-monotonic property that CCS possesses has very different semantics vis-a-vis the traditional support measure. In particular, "good" rules have a low CCS value. This makes CCS an ideal measure to use in conjunction with a top-down algorithm. Finally, the nature of CCS allows the pruning of rules without the setting of any threshold parameter! To the best of our knowledge this is the first threshold-free algorithm in association rule mining for classification.

Original language	English
Title of host publication	KDD 2006
Subtitle of host publication	Proceedings of the Twelfth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Pages	517-522
Number of pages	6
State	Published - 2006
Externally published	Yes
Event	KDD 2006: 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - Philadelphia, PA, United States Duration: Aug 20 2006 → Aug 23 2006

Publication series

Name	Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Volume	2006

Conference

Conference	KDD 2006: 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Country/Territory	United States
City	Philadelphia, PA
Period	08/20/06 → 08/23/06

Keywords

Association Rules Mining
Classification
Imbalanced Data sets
Parameter-free mining

Cite this

@inproceedings{bf1c66c87fad411ea7b032dcf8f2af76,

title = "CCCS: A top-down associative classifier for imbalanced class distribution",

abstract = "In this paper we propose CCCS, a new algorithm for classification based on association rule mining. The key innovation in CCCS is the use of a new measure, the {"}Complement Class Support (CCS){"} whose application results in rules which are guaranteed to be positively correlated. Furthermore, the anti-monotonic property that CCS possesses has very different semantics vis-a-vis the traditional support measure. In particular, {"}good{"} rules have a low CCS value. This makes CCS an ideal measure to use in conjunction with a top-down algorithm. Finally, the nature of CCS allows the pruning of rules without the setting of any threshold parameter! To the best of our knowledge this is the first threshold-free algorithm in association rule mining for classification.",

keywords = "Association Rules Mining, Classification, Imbalanced Data sets, Parameter-free mining",

author = "Bavani Arunasalam and Sanjay Chawla",

year = "2006",

language = "English",

isbn = "1595933395",

series = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

pages = "517--522",

booktitle = "KDD 2006",

note = "null ; Conference date: 20-08-2006 Through 23-08-2006",

}

Arunasalam, B & Chawla, S 2006, CCCS: A top-down associative classifier for imbalanced class distribution. in KDD 2006: Proceedings of the Twelfth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, vol. 2006, pp. 517-522, KDD 2006: 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Philadelphia, PA, United States, 08/20/06.

CCCS : A top-down associative classifier for imbalanced class distribution. / Arunasalam, Bavani; Chawla, Sanjay.

KDD 2006: Proceedings of the Twelfth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2006. p. 517-522 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; Vol. 2006).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - CCCS

AU - Arunasalam, Bavani

AU - Chawla, Sanjay

PY - 2006

Y1 - 2006

N2 - In this paper we propose CCCS, a new algorithm for classification based on association rule mining. The key innovation in CCCS is the use of a new measure, the "Complement Class Support (CCS)" whose application results in rules which are guaranteed to be positively correlated. Furthermore, the anti-monotonic property that CCS possesses has very different semantics vis-a-vis the traditional support measure. In particular, "good" rules have a low CCS value. This makes CCS an ideal measure to use in conjunction with a top-down algorithm. Finally, the nature of CCS allows the pruning of rules without the setting of any threshold parameter! To the best of our knowledge this is the first threshold-free algorithm in association rule mining for classification.

AB - In this paper we propose CCCS, a new algorithm for classification based on association rule mining. The key innovation in CCCS is the use of a new measure, the "Complement Class Support (CCS)" whose application results in rules which are guaranteed to be positively correlated. Furthermore, the anti-monotonic property that CCS possesses has very different semantics vis-a-vis the traditional support measure. In particular, "good" rules have a low CCS value. This makes CCS an ideal measure to use in conjunction with a top-down algorithm. Finally, the nature of CCS allows the pruning of rules without the setting of any threshold parameter! To the best of our knowledge this is the first threshold-free algorithm in association rule mining for classification.

KW - Association Rules Mining

KW - Classification

KW - Imbalanced Data sets

KW - Parameter-free mining

UR - http://www.scopus.com/inward/record.url?scp=33749576324&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:33749576324

SN - 1595933395

SN - 9781595933393

T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

SP - 517

EP - 522

BT - KDD 2006

Y2 - 20 August 2006 through 23 August 2006

ER -

CCCS: A top-down associative classifier for imbalanced class distribution

Abstract

Publication series

Conference

Keywords

Other files and links

Fingerprint

Cite this