Delegating Classifiers

C Ferri, PA Flach, J Hernandez-Orallo

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

Abstract

A sensible use of classifiers must be based on the estimated reliability of their predictions. A cautious classifier would delegate the difficult or uncertain predictions to other, possibly more specialised, classifiers. In this paper we analyse and develop this idea of delegating classifiers in a systematic way. First, we design a two-step scenario where a first classifier chooses which examples to classify and delegates the difficult examples to train a second classifier. Secondly, we present an iterated scenario involving an arbitrary number of chained classifiers. We compare these scenarios to classical ensemble methods, such as bagging and boosting. We show experimentally that our approach is not far behind these methods in terms of accuracy, but with several advantages: (i) improved efficiency, since each classifier learns from fewer examples than the previous one; (ii) improved comprehensibility, since each classification derives from a single classifier; and (iii) the possibility to simplify the overall multiclassifier by removing the parts that lead to delegation.
Translated title of the contributionDelegating Classifiers
Original languageEnglish
Title of host publicationUnknown
EditorsRuss Greineer, Dale Schuurmans
PublisherAssociation for Computing Machinery (ACM)
ISBN (Print)1581138385
Publication statusPublished - Jul 2004

Bibliographical note

Conference Proceedings/Title of Journal: Proceedings of the 21st International Conference on Machine Learning (ICML 2004)

Fingerprint

Dive into the research topics of 'Delegating Classifiers'. Together they form a unique fingerprint.

Cite this