Pathfinder 6: A Bayesian Network for Medical Diagnosis with 109 Nodes

Pathfinder 6 is a Bayesian network sample from the bnlearn repository, modeling medical diagnostic knowledge. The network contains 109 nodes, 195 arcs, and 72,079 parameters, with an average Markov blanket size of 3.82. It was created by D. Heckerman, E. Horwitz, and B. Nathwani for the Pathfinder Project, published in 1992.

Use Cases

• Benchmarking structure learning algorithms based on the known network topology of 109 nodes and 195 arcs.
• Evaluating probabilistic inference methods based on the network's 72,079 parameters.
• Studying the complexity of medical diagnostic reasoning based on the network's average Markov blanket size and degree.
• Teaching concepts of Bayesian networks and expert systems based on a well-documented, discrete, large-scale model.

Strengths

• Well-documented provenance with a clear citation to the original 1992 research paper.
• Defined network structure with specific metrics: 109 nodes, 195 arcs, and 72,079 parameters.
• Serves as a standard benchmark from a recognized repository (bnlearn).

Limitations

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

Provenance

Source

D. Heckerman, E. Horwitz, and B. Nathwani; bnlearn Bayesian Network Repository.

Collection Method

Constructed for the Pathfinder expert system project.

Time Range

null

Freshness

null

Geography

null