Hailfinder_8: A Bayesian Network for Severe Weather Forecasting

A Bayesian network with 56 nodes and 66 arcs, containing 2656 parameters for probabilistic weather forecasting. The model was authored by B. Abramson, J. Brown, W. Edwards, A. Murphy, and R. L. Winkler and is referenced in their 1996 paper 'Hailfinder: A Bayesian system for forecasting severe weather'.

Use Cases

• Benchmarking Bayesian network inference algorithms based on the described network structure.
• Studying probabilistic reasoning for severe weather events based on the model's domain.
• Teaching concepts of graphical models and parameter estimation using a documented real-world example.
• Developing decision support tools for meteorology based on the probabilistic forecasting framework.

Strengths

• Well-documented network structure with 56 nodes and 66 arcs.
• Clear academic provenance with a cited 1996 paper by the original authors.
• Specific model statistics provided, including 2656 parameters and an average Markov blanket size of 3.54.

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.
• Last update date is unknown; freshness unverified.

Provenance

Source

bnlearn Bayesian Network Repository

Collection Method

Model specification for a discrete Bayesian network.

Time Range

null

Freshness

unknown

Geography

null