Learning parameters of Bayesian networks from incomplete data via importance sampling

Authors:
Carsten Riggelsen
Affiliations:
Department of Information and Computing Sciences, Utrecht University, P.O. Box 80.098, 3508 TB Utrecht, The Netherlands
Venue:
International Journal of Approximate Reasoning
Year:
2006

Citing 7
Cited 4

Statistical analysis with missing data

Statistical analysis with missing data
A Bayesian Method for the Induction of Probabilistic Networks from Data

Machine Learning
The EM algorithm for graphical association models with missing data

Computational Statistics & Data Analysis - Special issue dedicated to Toma´sˇ Havra´nek
Learning Bayesian Networks: The Combination of Knowledge and Statistical Data

Machine Learning
WinBUGS – A Bayesian modelling framework: Concepts, structure, and extensibility

Statistics and Computing
Learning Bayesian networks from incomplete databases

UAI'97 Proceedings of the Thirteenth conference on Uncertainty in artificial intelligence
Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images

IEEE Transactions on Pattern Analysis and Machine Intelligence

Approximation Methods for Efficient Learning of Bayesian Networks

Proceedings of the 2008 conference on Approximation Methods for Efficient Learning of Bayesian Networks
Parameter learning in Bayesian network using semantic constraints of conversational feedback

PRICAI'10 Proceedings of the 11th Pacific Rim international conference on Trends in artificial intelligence
Bayesian networks for social modeling

SBP'11 Proceedings of the 4th international conference on Social computing, behavioral-cultural modeling and prediction
Calibrating bayesian network representations of social-behavioral models

SBP'10 Proceedings of the Third international conference on Social Computing, Behavioral Modeling, and Prediction

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present an algorithm for learning parameters of Bayesian networks from incomplete data. By using importance sampling we are able to assign a score to imputation proposals depending on the quality of such a proposal in combination with the observed data. This in effect makes it possible to approximate the posterior parameter distribution given incomplete data by using a mixture distribution with a tractable number of components. The technique allows for different imputation methods, in particular we propose an imputation method that combines Gibbs sampling and a data augmentation derivative. We evaluate our algorithm, and we compare the results to those obtained with WinBUGS and the EM algorithm.