A THEORY OF CROSS-VALIDATION ERROR

被引：15

作者：

TURNEY, P

机构：

[1] Knowledge Systems Laboratory, Institute for Information Technology, National Research Council Canada, Ottawa, ON

来源：

JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE | 1994年 / 6卷 / 04期

关键词：

CROSS-VALIDATION; SIMPLICITY; BIAS; VARIANCE; AIC; LINEAR REGRESSION; INSTANCE-BASED LEARNING;

D O I：

10.1080/09528139408953794

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 theory of error in cross-validation testing of algorithms for predicting real-valued attributes. The theory justifies the claim that predicting real-valued attributes requires balancing the conflicting demands of simplicity and accuracy. Furthermore, the theory indicates precisely how these conflicting demands must be balanced, in order to minimize cross-validation error. A general theory is presented, then it is developed in detail for linear regression and instance-based learning

引用

页码：361 / 391

页数：31

共 17 条

[1] INSTANCE-BASED LEARNING ALGORITHMS [J].

AHA, DW ;

KIBLER, D ;

ALBERT, MK .

MACHINE LEARNING, 1991, 6 (01) :37-66

[2]

AHA DW, 1989, 11TH P INT JOINT C A, P794

[3] NEW LOOK AT STATISTICAL-MODEL IDENTIFICATION [J].

AKAIKE, H .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1974, AC19 (06) :716-723

[4] STATISTICAL PREDICTOR IDENTIFICATION [J].

AKAIKE, H .

ANNALS OF THE INSTITUTE OF STATISTICAL MATHEMATICS, 1970, 22 (02) :203-&

[5]

Akaike H., 1992, 2 INT S INF THEOR, P267, DOI DOI 10.1007/978-1-4612-1694-0_15

[6]

Barron A.R., 1984, SELF ORG METHODS MOD, P87

[7]

Dasarathy B., 1991, NEAREST NEIGHBOR PAT

[8]

DRAPER NR, 1981, APPLIED REGRESSION A

[9] ATTRIBUTES OF THE PERFORMANCE OF CENTRAL PROCESSING UNITS - A RELATIVE PERFORMANCE PREDICTION MODEL [J].

EINDOR, P ;

FELDMESSER, J .

COMMUNICATIONS OF THE ACM, 1987, 30 (04) :308-317

[10]

Eubank R.L., 1988, SPLINE SMOOTHING NON

← 1 2 →