Statements (50)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:probabilistic_graphical_model
|
| gptkbp:alsoKnownAs |
gptkb:Bayes_nets
belief networks |
| gptkbp:appliesTo |
gptkb:diagnosis
gptkb:artificial_intelligence computer vision information retrieval speech recognition bioinformatics risk analysis |
| gptkbp:component |
conditional probability tables
edges (conditional dependencies) nodes (random variables) |
| gptkbp:introducedIn |
1980s
|
| gptkbp:mathematicalFoundation |
gptkb:Bayes'_theorem
|
| gptkbp:originatedIn |
gptkb:Judea_Pearl
|
| gptkbp:property |
can be constructed manually or learned from data
can be extended to dynamic Bayesian networks can be extended to influence diagrams can be used for anomaly detection can be used for causal discovery can be used for data fusion can be used for decision support can be used for knowledge representation can be used for reasoning under uncertainty can be used for time series analysis can encode conditional independence can handle missing data can represent joint probability distributions support efficient inference algorithms support exact and approximate inference support learning from data support parameter learning support structure learning |
| gptkbp:relatedTo |
gptkb:dynamic_Bayesian_networks
gptkb:hidden_Markov_models gptkb:Markov_networks causal networks |
| gptkbp:represents |
set of variables and their conditional dependencies
|
| gptkbp:usedFor |
gptkb:diagnosis
gptkb:machine_learning decision making probabilistic inference prediction causal reasoning |
| gptkbp:uses |
gptkb:directed_acyclic_graph
|
| gptkbp:bfsParent |
gptkb:Judea_Pearl
gptkb:directed_acyclic_graph |
| gptkbp:bfsLayer |
5
|
| https://www.w3.org/2000/01/rdf-schema#label |
Bayesian networks
|