Structural learning of mixed noisy-OR Bayesian networks

0581903 - ÚTIA 2024 RIV US eng J - Článek v odborném periodiku
Vomlel, Jiří - Kratochvíl, Václav - Kratochvíl, F.
Structural learning of mixed noisy-OR Bayesian networks.
International Journal of Approximate Reasoning. Roč. 161, č. 1 (2023), č. článku 108990. ISSN 0888-613X. E-ISSN 1873-4731
Grant CEP: GA ČR GA20-18407S
Institucionální podpora: RVO:67985556
Klíčová slova: Bayesian networks * Learning Bayesian networks * Linguistics * Loanwords
Obor OECD: Communication engineering and systems
Impakt faktor: 3.9, rok: 2022
Způsob publikování: Omezený přístup
http://library.utia.cas.cz/separaty/2024/MTR/vomlel-0581903.pdf https://www.sciencedirect.com/science/article/pii/S0888613X23001214?via%3Dihub

In this paper we discuss learning Bayesian networks whose conditional probability tables are either Noisy-OR models or general conditional probability tables. We refer to these models as Mixed Noisy-OR Bayesian Networks. To learn their structure, we modify the Bayesian Information Criterion used for standard Bayesian networks to reflect the number of parameters of a Noisy-OR model. We prove that the log-likelihood function of a Noisy-OR model has a unique maximum and adapt the EM-learning method for the leaky Noisy-OR model. We propose a structure learning algorithm that learns optimal Mixed Noisy-OR Bayesian Networks. We evaluate the proposed approach on synthetic data, where it performs substantially better than standard Bayesian networks. We perform experiments with Bipartite Noisy-OR Bayesian networks of different complexity to find out when the results of Mixed Noisy-OR Bayesian Networks are significantly better than the results of standard Bayesian networks and when they perform similarly. We also study how different penalties based on the number of model parameters affect the quality of the results. Finally, we apply the suggested approach to a problem from the domain of linguistics. Specifically, we use Mixed Noisy-OR Bayesian Networks to model the spread of loanwords in the South-East Asian Archipelago. We perform numerical experiments in which we compare the prediction ability of standard Bayesian networks with Mixed Noisy-OR Bayesian networks and test different pruning methods to reduce the number of parent sets considered.
Trvalý link: https://hdl.handle.net/11104/0350580