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

被引：101

作者：

Didaci, L ^{[1
]}

Giacinto, G ^{[1
]}

Roli, F ^{[1
]}

Marcialis, GL ^{[1
]}

机构：

[1] Univ Cagliari, Dept Elect & Elect Engn, I-09123 Cagliari, Italy

来源：

PATTERN RECOGNITION | 2005年 / 38卷 / 11期

关键词：

multiple classifier systems; dynamic classifier selection; performance evaluation;

D O I：

10.1016/j.patcog.2005.02.010

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

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. (c) 2005 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.

引用

页码：2188 / 2191

页数：4

共 5 条

[1] GIACINTO G, 1997, LECT NOTES COMPUTER, V1310, P38
[2] Giacinto G., 1999, P 10 INT C IM AN PRO, P659, DOI [10.1109/ICIAP.1999.797670, DOI 10.1109/ICIAP.1999.797670]
[3] Discriminant adaptive nearest neighbor classification
Hastie, T
Tibshirani, R
[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1996, 18 (06) : 607 - 616
[4] Switching between selection and fusion in combining classifiers: An experiment
Kuncheva, LI
[J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2002, 32 (02): : 146 - 156
[5] Combination of multiple classifiers using local accuracy estimates
Woods, K
Kegelmeyer, WP
Bowyer, K
[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1997, 19 (04) : 405 - 410

← 1 →