Combination of multiple classifiers using local accuracy estimates

被引：678

作者：

Woods, K ^{[1
]}

Kegelmeyer, WP ^{[1
]}

Bowyer, K ^{[1
]}

机构：

[1] SANDIA NATL LABS, CTR COMPUTAT ENGN, LIVERMORE, CA 94551 USA

来源：

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE | 1997年 / 19卷 / 04期

关键词：

combination of classifiers; dynamic classifier selection; local classifier accuracy; classifier fusion; ROC analysis;

D O I：

10.1109/34.588027

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper presents a method for combining classifiers that uses estimates of each individual classifier's local accuracy in small regions of feature space surrounding an unknown test sample. An empirical evaluation using five real data sets confirms the validity of our approach compared to some other Combination of Multiple Classifiers algorithms. We also suggest a methodology for determining the best mix of individual classifiers.

引用

页码：405 / 410

页数：6

共 18 条

[1]

[Anonymous], 1982, Pattern recognition: A statistical approach

[2]

Breiman L., 1984, Classification and Regression Trees, DOI DOI 10.2307/2530946

[3]

Duda R. O., 1973, PATTERN CLASSIFICATI, V3

[4] THE MEANING AND USE OF THE AREA UNDER A RECEIVER OPERATING CHARACTERISTIC (ROC) CURVE [J].

HANLEY, JA ;

MCNEIL, BJ .

RADIOLOGY, 1982, 143 (01) :29-36

[5]

HO TK, 1994, IEEE T PATTERN ANAL, V16, P66, DOI 10.1109/34.273716

[6] METHOD OF COMBINING MULTIPLE EXPERTS FOR THE RECOGNITION OF UNCONSTRAINED HANDWRITTEN NUMERALS [J].

HUANG, YS ;

SUEN, CY .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1995, 17 (01) :90-94

[7]

HUANG YS, 1994, INT C PATT RECOG, P473, DOI 10.1109/ICPR.1994.576986

[8]

Kegelmeyer W. P. Jr., 1994, V1069, P3

[9] HANDWRITTEN NUMERICAL RECOGNITION BASED ON MULTIPLE ALGORITHMS [J].

KIMURA, F ;

SHRIDHAR, M .

PATTERN RECOGNITION, 1991, 24 (10) :969-983

[10]

KNIGHT K, 1990, COMMUN ACM, V33, P59, DOI 10.1145/92755.92764

← 1 2 →