Experiments with real-life decision tables

Mikhail Ju Moshkov*, Marcin Piliszczuk, Beata Zielosko

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

2 Scopus citations

Abstract

This chapter is devoted to consideration of results of experiments with decision tables from UCI Repository of Machine Learning Databases [32]. The aim of the first group of experiments is to verify 0.5-hypothesis for real-life decision tables. We made experiments with 23 decision tables. Results of 21 experiments confirm 0.5-hypothesis for reducts: under the construction of partial test, during each step the greedy algorithm chooses an attribute which separates at least one-half of unseparated pairs of different rows with different decisions.

Original languageEnglish (US)
Title of host publicationPartial Covers, Reducts and Decision Rules in Rough Sets
Subtitle of host publicationTheory and Applications
EditorsMikhail Moshkov, Beata Zielosko, Marcin Piliszczuk
Pages117-133
Number of pages17
DOIs
StatePublished - 2008

Publication series

NameStudies in Computational Intelligence
Volume145
ISSN (Print)1860-949X

ASJC Scopus subject areas

  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Experiments with real-life decision tables'. Together they form a unique fingerprint.

Cite this