Importance and interaction of conditions in decision rules

Salvatore Greco, Benedetto Matarazzo, Roman Slowinski, Jerzy Stefanowski

Research output: Chapter in Book/Report/Conference proceedingChapter (peer-reviewed)peer-review

Abstract

Knowledge discovered from data tables is often presented in terms of "if... then..." decision rules. With each rule a confidence measure is associated. We present a method for measuring importance of each single condition or interactions among groups of conditions in the "if" part of the rules. The methodology is based on some indices introduced in literature to analyze fuzzy measures.
Original languageEnglish
Title of host publicationRough sets and current trends in computing
Subtitle of host publicationthird international conference, RSCTC 2002 Malvern, PA, USA, october 14–16, 2002 proceedings
EditorsJames J. Alpigini, James F. Peters, Andrzej Skowron, Ning Zhong
Place of PublicationBerlin, Germany
PublisherSpringer
Pages255-262
Number of pages8
Volume2475
ISBN (Print)9783540442745
DOIs
Publication statusPublished - Oct 2002

Publication series

NameLecture notes in computer science
PublisherSpringer
Number2475

Cite this