TY - JOUR
T1 - Hybrid Bayesian Networks Using Mixtures of Truncated Basis Functions
AU - Pérez-Bernabé, Inmaculada
AU - Maldonado, Ana D.
AU - Nielsen, Thomas D.
AU - Salmerón, Antonio
N1 - Funding Information:
This research has been partly funded by the Spanish Ministry of Science and Innovation, through projects TIN2016-77902-C3-3-P, PID2019-106758GB-C32 and by ERDF-FEDER funds.
Publisher Copyright:
© 2020
PY - 2020
Y1 - 2020
N2 - This paper introduces MoTBFs, an R package for manipulating mixtures of truncated basis functions. This class of functions allows the representation of joint probability distributions involving discrete and continuous variables simultaneously, and includes mixtures of truncated exponentials and mixtures of polynomials as special cases. The package implements functions for learning the parameters of univariate, multivariate, and conditional distributions, and provides support for parameter learning in Bayesian networks with both discrete and continuous variables. Probabilistic inference using forward sampling is also implemented. Part of the functionality of the MoTBFs package relies on the bnlearn package, which includes functions for learning the structure of a Bayesian network from a data set. Leveraging this functionality, the MoTBFs package supports learning of MoTBF-based Bayesian networks over hybrid domains. We give a brief introduction to the methodological context and algorithms implemented in the package. An extensive illustrative example is used to describe the package, its functionality, and its usage.
AB - This paper introduces MoTBFs, an R package for manipulating mixtures of truncated basis functions. This class of functions allows the representation of joint probability distributions involving discrete and continuous variables simultaneously, and includes mixtures of truncated exponentials and mixtures of polynomials as special cases. The package implements functions for learning the parameters of univariate, multivariate, and conditional distributions, and provides support for parameter learning in Bayesian networks with both discrete and continuous variables. Probabilistic inference using forward sampling is also implemented. Part of the functionality of the MoTBFs package relies on the bnlearn package, which includes functions for learning the structure of a Bayesian network from a data set. Leveraging this functionality, the MoTBFs package supports learning of MoTBF-based Bayesian networks over hybrid domains. We give a brief introduction to the methodological context and algorithms implemented in the package. An extensive illustrative example is used to describe the package, its functionality, and its usage.
UR - http://www.scopus.com/inward/record.url?scp=85101767787&partnerID=8YFLogxK
UR - https://journal.r-project.org/archive/2020/RJ-2021-019/RJ-2021-019.zip
U2 - 10.32614/rj-2021-019
DO - 10.32614/rj-2021-019
M3 - Journal article
AN - SCOPUS:85101767787
SN - 2073-4859
VL - 12
SP - 321
EP - 341
JO - R Journal
JF - R Journal
IS - 2
ER -