Details
| Original language | English |
|---|---|
| Pages (from-to) | 126-148 |
| Number of pages | 23 |
| Journal | Advances in Engineering Software |
| Volume | 115 |
| Early online date | 22 Sept 2017 |
| Publication status | Published - Jan 2018 |
Abstract
Bayesian Networks are a flexible and intuitive tool associated with a robust mathematical background. They have attracted increasing interest in a large variety of applications in different fields. In spite of this, inference in traditional Bayesian Networks is generally limited to only discrete variables or to probabilistic distributions (adopting approximate inference algorithms) that cannot fully capture the epistemic imprecision of the data available. In order to overcome these limitations, Credal Networks have been proposed to integrate Bayesian Networks with imprecise probabilities which, adopting non-probabilistic or hybrid models, allow to fully represent the information available and its uncertainty. Here, a novel computational tool, implemented in the general purpose software OpenCossan, is proposed. The tool provides the reduction of Credal Networks through the use of structural reliability methods, in order to limit the cost associated with the inference computation without impoverishing the quality of the information initially introduced. Novel algorithms for the inference computation of networks involving probability bounds are provided. In addition, a novel sensitivity approach is proposed and implemented into the Toolbox in order to identify the maximum tolerable uncertainty associated with the inputs.
Keywords
- Bayesian Networks, Credal Networks, Decision making, System reliability
ASJC Scopus subject areas
- Computer Science(all)
- Software
- Engineering(all)
- General Engineering
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Advances in Engineering Software, Vol. 115, 01.2018, p. 126-148.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - An open toolbox for the reduction, inference computation and sensitivity analysis of Credal Networks
AU - Tolo, Silvia
AU - Patelli, Edoardo
AU - Beer, Michael
N1 - Funding information: This work has been partially supported by the European Union’s Research and Innovation funding programme (Framework Programme) under the PLENOSE project (grant agreement number PIRSES-GA-2013-612581) and by the Newton Fund PhD Placement Programme.
PY - 2018/1
Y1 - 2018/1
N2 - Bayesian Networks are a flexible and intuitive tool associated with a robust mathematical background. They have attracted increasing interest in a large variety of applications in different fields. In spite of this, inference in traditional Bayesian Networks is generally limited to only discrete variables or to probabilistic distributions (adopting approximate inference algorithms) that cannot fully capture the epistemic imprecision of the data available. In order to overcome these limitations, Credal Networks have been proposed to integrate Bayesian Networks with imprecise probabilities which, adopting non-probabilistic or hybrid models, allow to fully represent the information available and its uncertainty. Here, a novel computational tool, implemented in the general purpose software OpenCossan, is proposed. The tool provides the reduction of Credal Networks through the use of structural reliability methods, in order to limit the cost associated with the inference computation without impoverishing the quality of the information initially introduced. Novel algorithms for the inference computation of networks involving probability bounds are provided. In addition, a novel sensitivity approach is proposed and implemented into the Toolbox in order to identify the maximum tolerable uncertainty associated with the inputs.
AB - Bayesian Networks are a flexible and intuitive tool associated with a robust mathematical background. They have attracted increasing interest in a large variety of applications in different fields. In spite of this, inference in traditional Bayesian Networks is generally limited to only discrete variables or to probabilistic distributions (adopting approximate inference algorithms) that cannot fully capture the epistemic imprecision of the data available. In order to overcome these limitations, Credal Networks have been proposed to integrate Bayesian Networks with imprecise probabilities which, adopting non-probabilistic or hybrid models, allow to fully represent the information available and its uncertainty. Here, a novel computational tool, implemented in the general purpose software OpenCossan, is proposed. The tool provides the reduction of Credal Networks through the use of structural reliability methods, in order to limit the cost associated with the inference computation without impoverishing the quality of the information initially introduced. Novel algorithms for the inference computation of networks involving probability bounds are provided. In addition, a novel sensitivity approach is proposed and implemented into the Toolbox in order to identify the maximum tolerable uncertainty associated with the inputs.
KW - Bayesian Networks
KW - Credal Networks
KW - Decision making
KW - System reliability
UR - http://www.scopus.com/inward/record.url?scp=85029700447&partnerID=8YFLogxK
U2 - 10.1016/j.advengsoft.2017.09.003
DO - 10.1016/j.advengsoft.2017.09.003
M3 - Article
AN - SCOPUS:85029700447
VL - 115
SP - 126
EP - 148
JO - Advances in Engineering Software
JF - Advances in Engineering Software
SN - 0965-9978
ER -