A study on the performances of dynamic classifier selection based on local accuracy estimation

by: Luca Didaci, Giorgio Giacinto, Fabio Roli, Gian L Marcialis

Pattern Recognition, Vol. 38, No. 11. (November 2005), pp. 2188-2191.

View FullText article

DOI, ScienceDirect

Reviews [Write a review of this article]

There are no reviews of this article

Find related articles from these CiteULike users

drivard

Find related articles with these CiteULike tags

classifier_ensemble, classifier_selection, oracle

Abstract

Dynamic classifier selection (DCS) plays a strategic role in the field of multiple classifier systems (MCS). This paper proposes a study on the performances of DCS by Local Accuracy estimation (DCS-LA). To this end, upper bounds against which the performances can be evaluated are proposed. The experimental results on five datasets clearly show the effectiveness of the selection methods based on local accuracy estimates.

@article{didaci_2005_pr, abstract = {Dynamic classifier selection (DCS) plays a strategic role in the field of multiple classifier systems (MCS). This paper proposes a study on the performances of DCS by Local Accuracy estimation (DCS-LA). To this end, upper bounds against which the performances can be evaluated are proposed. The experimental results on five datasets clearly show the effectiveness of the selection methods based on local accuracy estimates.}, author = {Didaci, Luca and Giacinto, Giorgio and Roli, Fabio and Marcialis, Gian L. }, citeulike-article-id = {2426383}, doi = {10.1016/j.patcog.2005.02.010}, journal = {Pattern Recognition}, keywords = {classifier\_ensemble, classifier\_selection, oracle}, month = {November}, number = {11}, pages = {2188--2191}, posted-at = {2008-02-25 18:33:28}, priority = {0}, title = {A study on the performances of dynamic classifier selection based on local accuracy estimation}, url = {http://dx.doi.org/10.1016/j.patcog.2005.02.010}, volume = {38}, year = {2005} }

RIS record

TY - JOUR ID - didaci_2005_pr L3 - citeulike-article-id:2426383 TI - A study on the performances of dynamic classifier selection based on local accuracy estimation JF - Pattern Recognition VL - 38 IS - 11 SP - 2188 EP - 2191 N2 - Dynamic classifier selection (DCS) plays a strategic role in the field of multiple classifier systems (MCS). This paper proposes a study on the performances of DCS by Local Accuracy estimation (DCS-LA). To this end, upper bounds against which the performances can be evaluated are proposed. The experimental results on five datasets clearly show the effectiveness of the selection methods based on local accuracy estimates. KW - classifier_ensemble KW - classifier_selection KW - oracle AU - Didaci, Luca AU - Giacinto, Giorgio AU - Roli, Fabio AU - Marcialis, Gian PY - 2005/11// UR - http://dx.doi.org/10.1016/j.patcog.2005.02.010 ER -

RIS BibTeX