Improving the practice of classifier performance assessment

被引:24
作者
Adams, NM [1 ]
Hand, DJ [1 ]
机构
[1] Univ London Imperial Coll Sci Technol & Med, Dept Math, London SW7 2BZ, England
关键词
D O I
10.1162/089976600300015808
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this note we use examples from the literature to illustrate some poor practices in assessing the performance of supervised classification rules, and we suggest guidelines for better methodology. We also describe a new assessment criterion that is suitable for the needs of many practical problems.
引用
收藏
页码:305 / 311
页数:7
相关论文
共 12 条
[1]   Comparing classifiers when the misallocation costs are uncertain [J].
Adams, NM ;
Hand, DJ .
PATTERN RECOGNITION, 1999, 32 (07) :1139-1147
[2]  
ADAMS NM, 2000, IN PRESS COMPUTERS B
[3]   The use of the area under the roc curve in the evaluation of machine learning algorithms [J].
Bradley, AP .
PATTERN RECOGNITION, 1997, 30 (07) :1145-1159
[4]  
Davison A. C., 1997, BOOTSTRAP METHODS TH, DOI 10.1017/CBO9780511802843
[5]   Prediction of outcome in critically ill patients using artificial neural network synthesised by genetic algorithm [J].
Dybowski, R ;
Weller, P ;
Chang, R ;
Gant, V .
LANCET, 1996, 347 (9009) :1146-1150
[6]  
Efron B., 1993, INTRO BOOTSTRAP, DOI 10.1007/978-1-4899-4541-9
[7]  
Hand DJ, 1997, J R STAT SOC A STAT, V160, P523
[8]  
HAND DJ, 1998, STAT FINANCE
[9]   Discrimination between the ulcerous and the nonulcerous forms of interstitial cystitis by noninvasive findings [J].
Koziol, JA ;
Adams, HP ;
Frutos, A .
JOURNAL OF UROLOGY, 1996, 155 (01) :87-90
[10]  
Michie D., 1994, Technometrics, V37, P459, DOI DOI 10.2307/1269742