Main content

Acquisition of Causal Models for Local Distributions in Bayesian Networks

Show simple item record Xiang, Yang Truong, Minh 2015-06-19T19:13:22Z 2015-06-19T19:13:22Z 2014
dc.identifier.citation Y. Xiang and M. Truong, Acquisition of Causal Models for Local Distributions in Bayesian Networks. IEEE Transactions on Cybernetics, Vol.44, No.9, 1591-1604, 2014. DOI: 10.1109/TCYB.2013.2290775 en_US
dc.identifier.issn 2168-2267
dc.description.abstract To specify a Bayesian network, a local distribution in the form of a conditional probability table, often of an effect conditioned on its n causes, needs to be acquired, one for each non-root node. Since the number of parameters to be assessed is generally exponential in n, improving the efficiency is an important concern in knowledge engineering. Non-impeding noisy-AND (NIN-AND) tree causal models reduce the number of parameters to being linear inn, while explicitly expressing both reinforcing and undermining interactions among causes. The key challenge in NIN-AND tree modeling is the acquisition of the NIN-AND tree structure. In this work, we formulate a concise structure representation and an expressive causal interaction function of NIN-AND trees. Building on these representations, we propose two structural acquisition methods, which are applicable to both elicitation-based and machine learning-based acquisitions. Their accuracy is demonstrated through experimental evaluations. en_US
dc.description.sponsorship NSERC, Canada en_US
dc.language.iso en en_US
dc.publisher IEEE en_US
dc.subject Bayesian networks, Local distributions, Causal independence, Knowledge acquisition, Elicitation, Graphical models, Noisy-OR models, Causal modeling, NIN-AND trees en_US
dc.title Acquisition of Causal Models for Local Distributions in Bayesian Networks en_US
dc.type Article en_US
dc.rights.license All items in the Atrium are protected by copyright with all rights reserved unless otherwise indicated.

Files in this item

Files Size Format View Description
cyber13.pdf 307.0Kb PDF View/Open Main article

This item appears in the following Collection(s)

Show simple item record