Change search
ReferencesLink to record
Permanent link

Direct link
On handling conflicts between rules with numerical features
KTH, School of Information and Communication Technology (ICT), Computer and Systems Sciences, DSV.
2006 (English)In: Proc ACM Symp Appl Computing, 2006, 37-41 p.Conference paper (Refereed)
Abstract [en]

Rule conflicts can arise in machine learning systems that utilise unordered rule sets. A rule conflict is when two or more rules cover the same example but differ in their majority classes. This conflict must be solved before a classification can be made. The standard methods for solving this type of problem are to use naive Bayes to solve the conflict or using the most frequent class (CN2). This paper studies the problem of rule conflicts in the area of numerical features. A novel family of methods, called distance based methods, for solving rule conflicts in continuous domains is presented. An empirical evaluation between a distance based method, CN2 and naive Bayes is made. It is shown that the distance based method significantly outperforms both naive Bayes and CN2.

Place, publisher, year, edition, pages
2006. 37-41 p.
, Proceedings of the ACM Symposium on Applied Computing, 1
Keyword [en]
Numerical features, Rule conflicts, Rule learning, Classification (of information), Numerical methods, Problem solving, Set theory, Learning systems
National Category
Computer Systems
URN: urn:nbn:se:kth:diva-155963ScopusID: 2-s2.0-33751052308ISBN: 1595931082ISBN: 9781595931085OAI: diva2:766147
2006 ACM Symposium on Applied Computing, 23-27 April 2006, Dijon, France

QC 20141126

Available from: 2014-11-26 Created: 2014-11-17 Last updated: 2014-11-26Bibliographically approved

Open Access in DiVA

No full text


Search in DiVA

By author/editor
Lindgren, Tony
By organisation
Computer and Systems Sciences, DSV
Computer Systems

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 3 hits
ReferencesLink to record
Permanent link

Direct link