Modeling the risk of structural fire incidents using a self-organizing map

摘要: Abstract A Self-Organizing Map (SOM) is used to classify and assess the risk levels of structural fire incidents. Such an assessment can be not only for improving safety protection existing future structures, but also enhancing emergency responses This reduce damages injuries resulting from fires. The map has a 2D hexagonal lattice structure was applied on sample incident records Toronto which were reported between 2000 2006. Assessment results suggest that SOM approach able successfully incidents with different properties into their predefined level classes. In summary, proposed shows superior performance predicting risk, although quality quantity training samples critical success predictions