Constructing Bayesian Networks Automatically using Ontologies
Year: 2006
Type: conference-paper
Abstract: Bayesian Networks are probabilistic structured representations of domains\nwhich have been applied to monitoring and manipulating cause and effects for modelled\nsystems as disparate as the weather, disease and mobile telecommunications networks.\nAlthough useful, Bayesian Networks are notoriously difficult to build accurately\nand efficiently which has somewhat limited their application to real world problems.\nOntologies are also a structured representation of knowledge, encoding facts and\nrules about a given domain. This paper outlines an approach to harness the knowledge\nand inference capabilities inherent in an ontology model to automate the construction\nof Bayesian Networks to accurately represent a domain of interest. The approach\nwas implemented in the context of an adaptive, self-configuring network management\nsystem in the telecommunications domain. In this system, the ontology model has\nthe dual function of knowledge repository and facilitator of automated workflows and\nthe generated BN serves to monitor effects of management activity, forming part of a\nfeedback look for self-configuration decisions and tasks.
Authors: Ann Devitt, Boris Danev, Katarína Matušíková
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2132450744
