Sequential optimization of γ-decision rules

Beata Zielosko

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

7 Scopus citations

Abstract

The paper is devoted to the study of an extension of dynamic programming approach which allows sequential optimization of approximate decision rules relative to length, coverage and number of misclassifications. Presented algorithm constructs a directed acyclic graph Δγ(T) which nodes are subtables of the decision table T. Based on the graph Δγ(T) we can describe all irredundant γ-decision rules with minimum length, after that among these rules describe all rules with maximum coverage, and among such rules describe all rules with minimum number of misclassifications. We can also change the set of cost functions and order of optimization. Sequential optimization can be considered as tool that help to construct simpler rules for understanding and interpreting by experts.
Original languageEnglish (US)
Title of host publication2012 Federated Conference on Computer Science and Information Systems (FedCSIS)
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
StatePublished - 2012

Bibliographical note

KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: The author would like to thank you Prof. Mikhail Moshkov, Dr. Igor Chikalov and Talha Amin for possibility to use results
of Dagger software system

Fingerprint

Dive into the research topics of 'Sequential optimization of γ-decision rules'. Together they form a unique fingerprint.

Cite this