A Bayesian network soil seismic liquefaction assessment approach considering physical prior knowledge under small dataset

Key Points

• The approach demonstrates effective prediction of liquefaction using a Bayesian network, which integrates physical prior knowledge.
• Findings indicate enhanced assessment of seismic effects in soils, particularly under conditions of limited data availability.
• Assessment utilizes a Bayesian network model, boosting accuracy in predicting liquefaction in various soil types.
• Improved methodologies may enable better preparedness for seismic events affecting vulnerable regions.