Rough Sets Based Minimal Certain Rule Generation in Non-deterministic Information Systems: An Overview

Rough sets based minimal certain rule generation in Non-deterministic Information Systems (NISs) is presented. This is advancement from rule generation in Deterministic Information Systems (DISs). NISs were proposed by Pawlak, Orlowska and Lipski in order to handle information incompleteness in DISs. According to the previous research on rule generation in DISs, rules in NISs are defined by logical implications satisfying some constraints. Rough sets based algorithms for generating rules are also proposed. Especially in this paper, minimal certain rules are focused on, and discernibility functions are newly introduced into NISs. A minimal certain rule is generated by means of a solution of a discernibility function. This procedure is implemented on a workstation in prolog.

[1]  Shusaku Tsumoto,et al.  Knowledge discovery in clinical databases and evaluation of discovered knowledge in outpatient clinic , 2000, Inf. Sci..

[2]  Marzena Kryszkiewicz,et al.  Rules in Incomplete Information Systems , 1999, Inf. Sci..

[3]  Zdzislaw Pawlak,et al.  Some Issues on Rough Sets , 2004, Trans. Rough Sets.

[4]  J. Kacprzyk,et al.  Incomplete Information: Rough Set Analysis , 1997 .

[5]  Hiroshi Sakai.  Some Effective Procedures for Data Dependencies in Information Systems , 2003 .

[6]  Witold Lipski,et al.  On Databases with Incomplete Information , 1981, JACM.

[7]  Hiroshi Sakai.  Effective Procedures for Handling Possible Equivalence Relations in Non-deterministic Information Systems , 2001, Fundam. Informaticae.

[8]  Jerzy W. Grzymala-Busse,et al.  Data with Missing Attribute Values: Generalization of Indiscernibility Relation and Rule Induction , 2004, Trans. Rough Sets.

[9]  Witold Lipski,et al.  On semantic issues connected with incomplete information databases , 1979, ACM Trans. Database Syst..

[10]  Marzena Kryszkiewicz,et al.  Computation of Reducts of Composed Information Systems , 1996, Fundam. Informaticae.

[11]  Hiroshi Sakai,et al.  Basic Algorithms and Tools for Rough Non-deterministic Information Analysis , 2004, Trans. Rough Sets.

[12]  Jerzy W. Grzymala-Busse,et al.  On the Best Search Method in the LEM1 and LEM2 Algorithms , 1998 .

[13]  Ewa Orlowska,et al.  Representation of Nondeterministic Information , 1984, Theor. Comput. Sci..