Bayesian Belief Network model data sets for seismic damage assessment of masonry buildings

This repository provides the datasets used to develop and validate a Bayesian Belief Network (BBN) model for seismic damage assessment of masonry buildings. The "Training_Database.xlsx" file contains observed data from post-earthquake surveys, including building characteristics, vulnerability parameters, seismic intensity measures, and corresponding damage levels, used to train the BBN model. The "Testing_Database.xlsx" file includes an independent dataset with the same variables, employed to evaluate the predictive performance and generalization capability of the model. The "Conditional_Probability_Tables.xlsx" file reports the conditional probability tables defining the probabilistic dependencies among variables in the BBN, enabling damage state inference under different seismic scenarios. These datasets support reproducibility and further research in probabilistic seismic risk assessment at both building and territorial scales.本仓库提供了用于开发与验证砌体建筑地震损伤评估贝叶斯信念网络（Bayesian Belief Network, BBN）模型的数据集。 文件"Training_Database.xlsx"收录了震后实地勘测所得的观测数据，涵盖建筑特征、易损性参数、地震强度度量指标及对应损伤等级，用于训练该贝叶斯信念网络模型。 文件"Testing_Database.xlsx"包含一组变量一致的独立数据集，用于评估该模型的预测性能与泛化能力。 文件"Conditional_Probability_Tables.xlsx"记录了定义贝叶斯信念网络内各变量间概率依赖关系的条件概率表，可支持不同地震场景下的损伤状态推演。 上述数据集可为建筑与区域尺度的概率地震风险评估领域的可复现研究及后续科研工作提供支撑。