win95pts_4: A 76-Node Bayesian Network Sample for Human-Computer Interaction

A sample from the Win95pts Bayesian Network, a discrete probabilistic model with 76 nodes and 112 arcs. The network contains 574 parameters and is referenced in the bnlearn Bayesian Network Repository. Its license is listed as us-pd.

Use Cases

• Structure learning and validation based on the 112 arcs and 76 nodes.
• Probabilistic inference and reasoning based on the 574 parameters.
• Benchmarking Bayesian network algorithms based on the average Markov blanket size of 5.92 and average degree of 2.95.
• Analyzing conditional independence relationships based on the graph structure.

Strengths

• The graph structure is explicitly defined with 76 nodes and 112 arcs.
• The model complexity is quantified with 574 parameters and a maximum in-degree of 7.
• Key network statistics are provided, including an average Markov blanket size of 5.92.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Sample data is unavailable for immediate inspection.

Provenance

Source

bnlearn Bayesian Network Repository (discrete-large section).

Collection Method

Likely a synthetic or simulated model representing a system.

Time Range

null

Freshness

Last update date is unknown; freshness unverified.

Geography

null