A 2-STAGE CLASSIFICATION SCHEME WITH BACKPROPAGATION NEURAL NETWORK CLASSIFIERS

被引：7

作者：

CHO, SB ^{[1
]}

KIM, JH ^{[1
]}

机构：

[1] KOREA ADV INST SCI & TECHNOL,DEPT COMP SCI,YOOSUNG KU,TAEJON 305701,SOUTH KOREA

来源：

PATTERN RECOGNITION LETTERS | 1992年 / 13卷 / 05期

关键词：

BACKPROPAGATION CLASSIFIER; 2-STAGE CLASSIFICATION;

D O I：

10.1016/0167-8655(92)90027-W

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 introduces backpropagation neural network as a classifier with some practical considerations, and proposes a two-stage classification scheme based on this classifier. A preliminary experiment with a large set of Hangul (Korean script) confirms the superiority relative to a single large neural network classifier.

引用

页码：309 / 313

页数：5

共 10 条

[1]

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

[2] A MULTISTAGE ALGORITHM FOR FAST CLASSIFICATION OF PATTERNS [J].

ELSHISHINY, H ;

ABDELMOTTALEB, MS ;

ELRAEY, M ;

SHOUKRY, A .

PATTERN RECOGNITION LETTERS, 1989, 10 (04) :211-215

[3] AN INTERACTIVE 2-LEVEL ARCHITECTURE FOR A MEMORY NETWORK PATTERN CLASSIFIER [J].

FAIRHURST, MC ;

WAHAB, HMSA .

PATTERN RECOGNITION LETTERS, 1990, 11 (08) :537-540

[4] ANALYSIS OF HIDDEN UNITS IN A LAYERED NETWORK TRAINED TO CLASSIFY SONAR TARGETS [J].

GORMAN, RP ;

SEJNOWSKI, TJ .

NEURAL NETWORKS, 1988, 1 (01) :75-89

[5] USING KEY FEATURES IN PATTERN-CLASSIFICATION [J].

KISSIOV, V ;

HADJITODOROV, S ;

KUNCHEVA, L .

PATTERN RECOGNITION LETTERS, 1990, 11 (01) :1-5

[6]

KOLMOGOROV AN, 1957, DOKL AKAD NAUK SSSR, V144, P679

[7] ON THE IDENTITY OF OPTIMAL STRATEGIES FOR MULTISTAGE CLASSIFIERS [J].

KURZYNSKI, MW .

PATTERN RECOGNITION LETTERS, 1989, 10 (01) :39-46

[8]

LIPPMANN RP, 1989, IEEE COMMUNICATI NOV, P47

[9]

Rumelhart DE, 1986, PARALLEL DISTRIBUTED, V1-2

[10]

Sejnowski T. J., 1987, Complex Systems, V1, P145

← 1 →