A prepositional satisfiability approach in mining compact rules

Azuraliza Abu Bakar, M. N. Sulaiman, M. Othman, M. H. Selamat

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

Abstract

This paper proposes a means of generating classification rules in rough classification theory by using a prepositional satisfiability (SAT) approach. The proposed method introduces a theoretical formalism that translates the discernibility relations of classes in a decision system into the SAT problem. The SAT representation is then formulated as an Integer Programming (IP) model called SAT_IP to solve the minimal reduct computation problem. First, a set of rules is generated from the reducts and these will then be used to solve the classification problem. The method generates a minimum set of rules, which are shorter in length, and there is also greater accuracy in the classification.

Original languageEnglish
Title of host publicationManagement Information Systems
EditorsA. Zanasi, C.A. Brebbia, N.F.F.E. Ebecken, P. Melli
PublisherWITPress
Pages197-204
Number of pages8
Volume6
ISBN (Print)1853128309
Publication statusPublished - 2002
Externally publishedYes
EventThird International Conference on Data Mining, Data Mining III - Bologna
Duration: 25 Sep 200227 Sep 2002

Other

OtherThird International Conference on Data Mining, Data Mining III
CityBologna
Period25/9/0227/9/02

Fingerprint

Integer programming

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Abu Bakar, A., Sulaiman, M. N., Othman, M., & Selamat, M. H. (2002). A prepositional satisfiability approach in mining compact rules. In A. Zanasi, C. A. Brebbia, N. F. F. E. Ebecken, & P. Melli (Eds.), Management Information Systems (Vol. 6, pp. 197-204). WITPress.

A prepositional satisfiability approach in mining compact rules. / Abu Bakar, Azuraliza; Sulaiman, M. N.; Othman, M.; Selamat, M. H.

Management Information Systems. ed. / A. Zanasi; C.A. Brebbia; N.F.F.E. Ebecken; P. Melli. Vol. 6 WITPress, 2002. p. 197-204.

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

Abu Bakar, A, Sulaiman, MN, Othman, M & Selamat, MH 2002, A prepositional satisfiability approach in mining compact rules. in A Zanasi, CA Brebbia, NFFE Ebecken & P Melli (eds), Management Information Systems. vol. 6, WITPress, pp. 197-204, Third International Conference on Data Mining, Data Mining III, Bologna, 25/9/02.
Abu Bakar A, Sulaiman MN, Othman M, Selamat MH. A prepositional satisfiability approach in mining compact rules. In Zanasi A, Brebbia CA, Ebecken NFFE, Melli P, editors, Management Information Systems. Vol. 6. WITPress. 2002. p. 197-204
Abu Bakar, Azuraliza ; Sulaiman, M. N. ; Othman, M. ; Selamat, M. H. / A prepositional satisfiability approach in mining compact rules. Management Information Systems. editor / A. Zanasi ; C.A. Brebbia ; N.F.F.E. Ebecken ; P. Melli. Vol. 6 WITPress, 2002. pp. 197-204
@inproceedings{53f612db0a724e5aa7687ea8ab7b6654,
title = "A prepositional satisfiability approach in mining compact rules",
abstract = "This paper proposes a means of generating classification rules in rough classification theory by using a prepositional satisfiability (SAT) approach. The proposed method introduces a theoretical formalism that translates the discernibility relations of classes in a decision system into the SAT problem. The SAT representation is then formulated as an Integer Programming (IP) model called SAT_IP to solve the minimal reduct computation problem. First, a set of rules is generated from the reducts and these will then be used to solve the classification problem. The method generates a minimum set of rules, which are shorter in length, and there is also greater accuracy in the classification.",
author = "{Abu Bakar}, Azuraliza and Sulaiman, {M. N.} and M. Othman and Selamat, {M. H.}",
year = "2002",
language = "English",
isbn = "1853128309",
volume = "6",
pages = "197--204",
editor = "A. Zanasi and C.A. Brebbia and N.F.F.E. Ebecken and P. Melli",
booktitle = "Management Information Systems",
publisher = "WITPress",

}

TY - GEN

T1 - A prepositional satisfiability approach in mining compact rules

AU - Abu Bakar, Azuraliza

AU - Sulaiman, M. N.

AU - Othman, M.

AU - Selamat, M. H.

PY - 2002

Y1 - 2002

N2 - This paper proposes a means of generating classification rules in rough classification theory by using a prepositional satisfiability (SAT) approach. The proposed method introduces a theoretical formalism that translates the discernibility relations of classes in a decision system into the SAT problem. The SAT representation is then formulated as an Integer Programming (IP) model called SAT_IP to solve the minimal reduct computation problem. First, a set of rules is generated from the reducts and these will then be used to solve the classification problem. The method generates a minimum set of rules, which are shorter in length, and there is also greater accuracy in the classification.

AB - This paper proposes a means of generating classification rules in rough classification theory by using a prepositional satisfiability (SAT) approach. The proposed method introduces a theoretical formalism that translates the discernibility relations of classes in a decision system into the SAT problem. The SAT representation is then formulated as an Integer Programming (IP) model called SAT_IP to solve the minimal reduct computation problem. First, a set of rules is generated from the reducts and these will then be used to solve the classification problem. The method generates a minimum set of rules, which are shorter in length, and there is also greater accuracy in the classification.

UR - http://www.scopus.com/inward/record.url?scp=2942631660&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2942631660&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:2942631660

SN - 1853128309

VL - 6

SP - 197

EP - 204

BT - Management Information Systems

A2 - Zanasi, A.

A2 - Brebbia, C.A.

A2 - Ebecken, N.F.F.E.

A2 - Melli, P.

PB - WITPress

ER -