Win95pts: Bayesian Network with 76 Nodes and 112 Arcs

A Bayesian network with 76 nodes and 112 arcs, containing 574 parameters. The structure, with an average Markov blanket size of 5.92 and a maximum in-degree of 7, is a discrete, large-scale model from the bnlearn repository. It is hosted on OpenML and is in the public domain in the United States.

Use Cases

• Benchmarking structure learning algorithms based on the known network topology.
• Testing probabilistic inference methods based on the 574 model parameters.
• Studying the properties of discrete graphical models based on the reported average degree and Markov blanket size.
• Comparing causal discovery techniques based on the directed arc structure.

Strengths

• Well-defined network structure with 76 nodes and 112 arcs.
• Known topological metrics, including an average Markov blanket size of 5.92 and a maximum in-degree of 7.
• Public domain license (us-pd) allows for unrestricted use.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for certain tasks.
• Last update date is unknown; freshness unverified.

Provenance

Source

bnlearn Bayesian Network Repository

Collection Method

Model specification from a known repository.

Time Range

null

Freshness

Last updated date is unknown.

Geography

null