Association models for web mining

被引:15
作者
Giudici, P
Castelo, R
机构
[1] Univ Pavia, Dipartimento Econ Polit & Metodi Quantitat, Fac Econ, I-27100 Pavia, Italy
[2] Univ Utrecht, NL-3508 TC Utrecht, Netherlands
关键词
Bayesian inference; data mining; graphical models; Markov chain Monte Carlo methods; model selection;
D O I
10.1023/A:1011469000311
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We describe how statistical association models and, specifically, graphical models, can be usefully employed to model web mining data. We describe some methodological problems related to the implementation of discrete graphical models for web mining data. In particular, we discuss model selection procedures.
引用
收藏
页码:183 / 196
页数:14
相关论文
共 9 条
[1]  
[Anonymous], 1998, P 14 C UNC ART INT
[2]  
Brooks SP, 1998, J ROY STAT SOC D-STA, V47, P69, DOI 10.1111/1467-9884.00117
[3]   Decomposable graphical Gaussian model determination [J].
Giudici, P ;
Green, PJ .
BIOMETRIKA, 1999, 86 (04) :785-801
[4]  
GIUDICI P, 2001, UNPUB IMPROVING MARK
[5]   Dependency networks for inference, collaborative filtering, and data visualization [J].
Heckerman, D ;
Chickering, DM ;
Meek, C ;
Rounthwaite, R ;
Kadie, C .
JOURNAL OF MACHINE LEARNING RESEARCH, 2001, 1 (01) :49-75
[6]  
HECKERMAN D, 1995, MACH LEARN, V20, P197, DOI 10.1007/BF00994016
[7]  
Lauritzen S.L., 1996, Graphical models, V17, P298
[8]   BAYESIAN GRAPHICAL MODELS FOR DISCRETE-DATA [J].
MADIGAN, D ;
YORK, J .
INTERNATIONAL STATISTICAL REVIEW, 1995, 63 (02) :215-232
[9]  
Whittaker J, 1990, GRAPHICAL MODELS APP