Abstract | ||
---|---|---|
Bayesian Networks (BNs) have been extensively used for diagnosis applications. Knowledge acquisition (KA), i.e. building a BN from the knowledge of experts in the application domain, involves two phases: knowledge gathering and model construction, i.e. defining the model based on that knowledge. The number of parameters involved in a large network is normally intractable to be specified by human experts. This leads to a trade-off between the accuracy of a detailed model and the size and complexity of such a model. In this paper, a Knowledge Acquisition Tool (KAT) to automatically perform information gathering and model construction for diagnosis of the radio access part of cellular networks is presented. KAT automatically builds a diagnosis model based on the experts’ answers to a sequence of questions regarding his way of reasoning in diagnosis. This will be performed for two BN structures: Simple Bayes Model (SBM) and Independence of Causal Influence (ICI) models. |
Year | DOI | Venue |
---|---|---|
2006 | 10.1007/11811220_6 | KSEM |
Keywords | Field | DocType |
causal influence,knowledge gathering,bayesian networks,detailed model,information gathering,bn structure,diagnosis application,diagnosis model,bayesian network,knowledge acquisition,model construction,cellular network | Data mining,Computer science,Expert system,Bayesian network,Artificial intelligence,Application domain,Knowledge engineering,Knowledge base,Radio access network,Knowledge acquisition,Machine learning,Bayes' theorem | Conference |
Volume | ISSN | ISBN |
4092.0 | 0302-9743 | 3-540-37033-1 |
Citations | PageRank | References |
1 | 0.39 | 15 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Raquel Barco | 1 | 364 | 41.12 |
Pedro Lazaro | 2 | 50 | 4.18 |
Volker Wille | 3 | 130 | 13.37 |
L. Díez | 4 | 141 | 23.21 |