SEQUENTIAL MODEL CRITICISM IN PROBABILISTIC EXPERT SYSTEMS

被引：27

作者：

COWELL, RG

DAWID, AP

SPIEGELHALTER, DJ

机构：

[1] MRC,BIOSTAT UNIT,CAMBRIDGE,ENGLAND

[2] CITY UNIV LONDON,LONDON EC1V 0HB,ENGLAND

来源：

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE | 1993年 / 15卷 / 03期

关键词：

D O I：

10.1109/34.204903

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Probabilistic expert systems based on Bayesian networks require initial specification of both qualitative graphical structure and quantitative conditional probability assessments. As (possibly incomplete) data accumulate on real cases, the parameters of the system may adapt, but it is also essential that the initial specifications are monitored with respect to their predictive performance. We present a range of monitors based on standardized scoring rules that are designed to detect both qualitative and quantitative departures from the specified model. A simulation study demonstrates the efficacy these monitors at uncovering such departures.

引用

页码：209 / 219

页数：11

共 26 条

[11] MAXIMUM LIKELIHOOD FROM INCOMPLETE DATA VIA EM ALGORITHM [J].

DEMPSTER, AP ;

LAIRD, NM ;

RUBIN, DB .

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL, 1977, 39 (01) :1-38

[12] A FAST MODEL SELECTION PROCEDURE FOR LARGE FAMILIES OF MODELS [J].

EDWARDS, D ;

HAVRANEK, T .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1987, 82 (397) :205-213

[13]

Glymour C., 1987, DISCOVERING CAUSAL S

[14]

Jensen F. V., 1990, Computational Statistics Quarterly, V5, P269

[15]

LAURITZEN SL, 1988, J ROY STAT SOC B MET, V50, P157

[16]

LAURITZEN SL, 1991, R9105 AALB U I EL SY

[17] PROBABILITY FORECASTING IN METEOROLOGY [J].

MURPHY, AH ;

WINKLER, RL .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1984, 79 (387) :489-500

[18]

PEARL J, 1991, 2ND P INT C PRINC KN, P441

[19]

Pearl J, 1988, PROBABILISTIC INFERE

[20]

SEILLIERMOISEIW.F, IN PRESS J AM STAT A

← 1 2 3 →