J4

• 计算机科学 • Previous Articles     Next Articles

An algorithm for discovery default rules based on GDT

QI Hong1, LIU Da-you1, LIU Ya-bo1, CUI Gu ang-cai1,2   

  1. 1. College of Computer Science and Technology, Jilin University, Changchun 130012, China; 2. College of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130022, China
  • Received:2003-11-24 Revised:1900-01-01 Online:2004-07-26 Published:2004-07-26
  • Contact: QI Hong

Abstract: By using a generalization distribution table (GDT) as a hypothesis search space, this paper presents an algorithm to discover default rules from the information system with inconsistencies and incompleteness, by which the default rule can be mined and the noisy in the data can be filtered, and unseen instances can be predicted.

Key words: generalization distribution table, default rules, inc onsistency, incompleteness

CLC Number: 

  • TP311