Automated Discovery of Discriminant Rules for a Group of Objects in Databases (1998) [2 citations — 2 self]
Popular Tags
No tags have been applied to this document.
BibTeX | Add To MetaCart
author = {Tae-Wan Ryu and Christoph F. Eick},
title = {Automated Discovery of Discriminant Rules for a Group of Objects in Databases},
year = {1998}
}
Bookmarks
OpenURL
Abstract:
This paper discusses our research overview and problems in discovering useful knowledge from a structured database. This research is an extended work of a prototype system, MASSON that discovers discriminant rules in the form of database query for a set of objects using genetic programming. In order to discover useful discriminant rules in a database, the basic approach that is employed in this research is first cluster the given database into clusters that contain a set of similar objects, then the set of objects in each cluster is provided to the discovery system, MASSON. We also discuss several research problems in clustering a structured database. Introduction Discriminant rules are rules that exclusively describe the characteristics for a set of objects in a given data set. In this framework, the rules cover only the specific set of objects without covering any other objects in a given data set. Discovery (or unsupervised learning) systems can discover such rules by themselves wh...
