Rule induction with extension matrices

✍ Scribed by Wu, Xindong

• Publisher: John Wiley and Sons
• Year: 1998
• Tongue: English
• Weight: 212 KB
• Volume: 49
• Category: Article
• ISSN: 0002-8231
• DOI: 10.1002/(sici)1097-4571(19980415)49:5<435::aid-asi6>3.0.co;2-r

✦ Synopsis

This article presents a heuristic, attribute-based, noise-knowledge discovery in databases (KDD), has been seen tolerant data mining program, HCV (Version 2.0), based (Michie, 1987;Quinlan, 1988; Wu, 1995) as not only a on the newly-developed extension matrix approach. By feasible way but also the only way of avoiding the knowldividing the positive examples (PE) of a specific class edge bottleneck problem. While it is often difficult for an in a given example set into intersecting groups and expert to articulate their expertise explicitly and clearly, adopting a set of strategies to find a heuristic conjunctive formula in each group which covers all the group's it is usually relatively easy to document case studies of positive examples and none of the negative examples their skill at work. The learning systems in commercial (NE), the HCV induction algorithm adopted in the HCV use today are almost exclusively inductive ones. The most (Version 2.0) software finds a description formula in the widespread family of learning algorithms for learning sysform of variable-valued logic for PE against NE in lowtems is the decision tree based ID3-like family (Cestnik, order polynomial time at induction time. In addition to the HCV induction algorithm, this article also outlines