General inference algorithm of Bayesian networks based on clique tree

Haijun Li; Xiao Liu

doi:10.1117/12.806449

13 October 2008 General inference algorithm of Bayesian networks based on clique tree

Haijun Li, Xiao Liu

Proceedings Volume 7128, Seventh International Symposium on Instrumentation and Control Technology: Measurement Theory and Systems and Aeronautical Equipment; 71280D (2008) https://doi.org/10.1117/12.806449
Event: Seventh International Symposium on Instrumentation and Control Technology, 2008, Beijing, China

Abstract

A general inference algorithm which based on exact algorithm of clique tree and importance sampling principle was put forward this article. It applied advantages of two algorithms, made information transfer from one clique to another, but don't calculate exact interim result. It calculated and dealt with the information using approximate algorithm, calculated the information from one clique to another using current potential. Because this algorithm was an iterative course of improvement, this continuous ran could increases potential of each clique, and produced much more exact information. Hybrid Bayesian Networks inference algorithm based on general softmax function could deal whit any function for CPD, and could be applicable for any models. Simulation test proved that the effect of classification was fine.

Citation Download Citation

Haijun Li and Xiao Liu "General inference algorithm of Bayesian networks based on clique tree", Proc. SPIE 7128, Seventh International Symposium on Instrumentation and Control Technology: Measurement Theory and Systems and Aeronautical Equipment, 71280D (13 October 2008); https://doi.org/10.1117/12.806449

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available