Generation of reducts and rules in multi-attribute and multi-criteria classification

The paper addresses the problem of analysing information tables which contain objects described by both attributes and criteria, i.e. attributes with preference-ordered scales. The objects contained in those tables, representing exemplary decisions made by a decision maker or a domain expert, are usually classified into one of several classes that are also often preference-ordered. Analysis of such data using the classic rough set methodology may produce improper results, as the original rough set approach is not able to discover inconsistencies originating from consideration of typical criteria, like e.g. product quality, market share or debt ratio. The paper presents the framework for the analysis of both attributes and criteria and a very promising algorithm for generating reducts. The algorithm presented is evaluated in an experiment with real-life data sets and its results are compared to those by two other reduct generating algorithms.

[1]  Robert Susmaga.  New test for inclusion minimality in reduct generation , 2000 .

[2]  D. Vanderpooten.  Similarity Relation as a Basis for Rough Approximations , 1995 .

[3]  Constantin Zopounidis,et al.  Application of the Rough Set Approach to Evaluation of Bankruptcy Risk , 1995 .

[4]  Daniel Vanderpooten,et al.  A General Two-Stage Approach to Inducing Rules from Examples , 1993, RSKD.

[5]  Roman Słowiński,et al.  A New Rough Set Approach to Evaluation of Bankruptcy Risk , 1998 .

[6]  Salvatore Greco,et al.  A New Rough Set Approach to Multicriteria and Multiattribute Classification , 1998, Rough Sets and Current Trends in Computing.

[7]  Thorsten Kuhlmann,et al.  Intelligent decision support , 1998 .

[8]  Salvatore Greco,et al.  Rough approximation by dominance relations , 2002, Int. J. Intell. Syst..

[9]  Roman Słowiński,et al.  The Use of Rough Sets and Fuzzy Sets in MCDM , 1999 .

[10]  Constantin Zopounidis,et al.  A survey of business failures with an emphasis on prediction methods and industrial applications , 1996 .

[11]  Stanislaw Romanski,et al.  Operations on Families of Sets for Exhaustive Search, Given a Monotonic Function , 1988, JCDKB.

[12]  J. Stefanowski,et al.  Rough Sets as a Tool for Studying Attribute Dependencies in the Urinary Stones Treatment Data Set , 1997 .

[13]  Janusz Zalewski,et al.  Rough sets: Theoretical aspects of reasoning about data , 1996 .

[14]  Jerzy W. Grzymala-Busse,et al.  Global discretization of continuous attributes as preprocessing for machine learning , 1996, Int. J. Approx. Reason..

[15]  Robert Susmaga.  Computation of Minimal Cost Reducts , 1999, ISMIS.

[16]  Roman Slowinski,et al.  Sensitivity Analysis of Rough Classification , 1990, Int. J. Man Mach. Stud..

[17]  Andrzej Skowron,et al.  The Discernibility Matrices and Functions in Information Systems , 1992, Intelligent Decision Support.

[18]  Robert Susmaga,et al.  Effective tests for minimality in reduct generation , 1998 .

[19]  S. Greco,et al.  Rough Approximation of a Preference Relation in a Pairwise Comparison Table , 1998 .