A note on comparing classifiers

被引:61
作者
Duin, RPW [1 ]
机构
[1] DELFT UNIV TECHNOL, FAC APPL PHYS, PATTERN RECOGNIT GRP, 2600 GA DELFT, NETHERLANDS
关键词
automatic classifiers; benchmarking; comparisons; feedforward neural networks;
D O I
10.1016/0167-8655(95)00113-1
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Recently many new classifiers have been proposed, mainly based on neural network techniques. Comparisons are needed to evaluate the performance of the new methods. It is argued that a straightforward fair comparison demands automatic classifiers with no user interaction. As this conflicts with one of the main characteristics of neural networks, their flexibility, the question whether they are better or worse than traditional techniques might be undecidable.
引用
收藏
页码:529 / 536
页数:8
相关论文
共 22 条
[1]   NEURAL NETWORKS - A REVIEW FROM A STATISTICAL PERSPECTIVE [J].
CHENG, B ;
TITTERINGTON, DM .
STATISTICAL SCIENCE, 1994, 9 (01) :2-30
[2]  
Cox L., 1982, ASA P STAT COMPUTATI, P55
[4]  
DUIN RPW, 1976, IEEE T COMPUT, V25, P1175, DOI 10.1109/TC.1976.1674577
[5]  
DUIN RPW, 1995, SCIA 95 P 9 SCAND C, V2, P957
[6]   A linkage test with mice [J].
Fisher, RA ;
Mather, K .
ANNALS OF EUGENICS, 1936, 7 :265-280
[7]  
Golomb B., 1991, Advances in Neural Information Processing Systems, V3, P572, DOI DOI 10.1007/978-1-4757-2379-3_3
[8]   LEARNED CLASSIFICATION OF SONAR TARGETS USING A MASSIVELY PARALLEL NETWORK [J].
GORMAN, RP ;
SEJNOWSKI, TJ .
IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1988, 36 (07) :1135-1140
[9]  
HOEKSTRA A, 1995, CONFIDENCE VALUE EST
[10]  
*ICMS, 1995, ICMS WORKSH STAT NEU