Improving modular classification rule induction with G-Prism using dynamic rule term boundaries

M. Almutairi, F. Stahl, Max Bramer

Research output: Chapter in Book/Report/Conference proceedingConference contribution

88 Downloads (Pure)

Abstract

Modular classification rule induction for predictive analytics is an alternative and expressive approach to rule induction as opposed to decision tree based classifiers. Prism classifiers achieve a similar classification accuracy compared with decision trees, but tend to overfit less, especially if there is noise in the data. This paper describes the development of a new member of the Prism family, the G-Prism classifier, which improves the classification performance of the classifier. G-Prism is different compared with the remaining members of the Prism family as it follows a different rule term induction strategy. G-Prism’s rule term induction strategy is based on Gauss Probability Density Distribution (GPDD) of target classes rather than simple binary splits (local discretisation). Two versions of G-Prism have been developed, one uses fixed boundaries to build rule terms from GPDD and the other uses dynamic rule term boundaries. Both versions have been compared empirically against Prism on 11 datasets using various evaluation metrics. The results show that in most cases both versions of G-Prism, especially G-Prism with dynamic boundaries, achieve a better classification performance compared with Prism.
Original languageEnglish
Title of host publicationArtificial Intelligence XXXIV
Subtitle of host publication37th SGAI International Conference on Artificial Intelligence, AI 2017, Cambridge, UK, December 12-14, 2017, Proceedings
EditorsMax Bramer, Miltos Petridis
PublisherSpringer
Pages115-128
ISBN (Electronic)978-3-319-71078-5
ISBN (Print)978-3-319-71077-8
DOIs
Publication statusPublished - Nov 2017
Event37th SGAI International Conference on Artificial Intelligence - Cambridge, United Kingdom
Duration: 12 Dec 201714 Dec 2017

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume10630
ISSN (Print)0302-9743

Conference

Conference37th SGAI International Conference on Artificial Intelligence
Abbreviated titleAI 2017
Country/TerritoryUnited Kingdom
CityCambridge
Period12/12/1714/12/17

Fingerprint

Dive into the research topics of 'Improving modular classification rule induction with G-Prism using dynamic rule term boundaries'. Together they form a unique fingerprint.

Cite this