Credibility coefficients reflect similarity of obj ects in respect to other ones in information systems. For decision tables we can use credibility coefficients based on decision rules. Knowledge discovery methods can extract rule s from an information system. The knowledge represented by the rules may be not exact due to improper data. Calculation of credibility coefficients is based on an assumption that majority of data is correct and only a minor part may be improper. The main purpose of u sing credibility coefficients is to indicate to which group a particular object probably belongs . A main focus of the paper is set on an algorithm of calculating credibility coefficients a nd a presentation how credibility coefficients can be used. The algorithm of presented credibility coefficients is based on decision rules, which are generated using the rough set theory. Some rema rks on practical results of identifying improper data by credibility coefficients are inser ted in the paper as well.
[1]
Roman Podraza,et al.
Credibility Coefficients in ARES Rough Set Exploration System
,
2005,
RSFDGrC.
[2]
Z. Pawlak.
Rough Sets: Theoretical Aspects of Reasoning about Data
,
1991
.
[3]
Roman Podraza,et al.
KTDA: emerging patterns based data analysis system
,
2006,
Ann. UMCS Informatica.
[4]
Jerzy W. Grzymala-Busse,et al.
Rough Sets
,
1995,
Commun. ACM.
[5]
R. Podraza,et al.
Credibility Coefficients for Objects of Rough Sets
,
2006
.
[6]
Uniwersytet Marii Curie-Skłodowskiej w Lublinie,et al.
Annales Universitatis Mariae Curie-Skłodowska
,
2001
.
[7]
Janusz Zalewski,et al.
Rough sets: Theoretical aspects of reasoning about data
,
1996
.
[8]
Andrzej Skowron,et al.
EXTRACTING LAWS FROM DECISION TABLES: A ROUGH SET APPROACH
,
1995,
Comput. Intell..