Search

The complexes of historic timber-structure buildings are extremely complicated and have various fire threats. A targeted approach to fire risk assessment of historic timber structures is conducive to focused fire plans. Due to the sparsity of the historical fire event dataset, it is quite challenging to predict the fire risk of heritage buildings based on historical fire events data. In this study, a fire risk index-driven solution is proposed based on machine learning to predict the fire risk level of timber heritage buildings. The suggested fire risk index system emphasizes the specificities of timber and historic structures, and the indicators are processed on temporal, spatial, and attribute scales. Considering the uncertainty of fire risk, this approach objectively rates the historical fire risk of heritage buildings using entropy-weighted TOPSIS. The typical machine learning model, XGBoost, is enhanced by applying the spatially unique values to predict the monthly fire risk level of a single heritage building. It solves the issue that general models only take into account the attribute characteristics of heritage buildings. Through the proposed solution, the monthly fire risk level of the heritage buildings of the Palace Museum of China in 2019 was assessed by the XGBoost model using the fire risk index data of the historical years (2016–2018). The accuracy evaluation showed that XGBoost performed better than other models (GBDT, CatBoost, Adaboost, and SVM), which is expected to provide information support for the safety inspection work of the fire department in the Palace Museum. [ABSTRACT FROM AUTHOR]

Some scholars have shown that the machine learning methods based on a single-source data can successfully monitor the risks of formal financial activities, but not those of informal financial activities. This is because the data generated by formal financial activities, whether it is the structured or unstructured data, are of high quality and quantity, while the data generated by informal financial activities are not. Therefore, multi-source data are the key to monitor the risks of informal financial activities through machine learning. Although a few studies attempted to use multi-source data for financial risk prediction, they simply stack the obtained multi-source data, but ignore the original sources, heterogeneity, mutual redundancy and other characteristics of the data, so that the improvement of the prediction effect is not obvious. Therefore, TSAIB_RS method based on the two-stage adaptive integration of multi-source heterogeneous data was constructed in the paper, in which the data with different sources and different distributions were adaptively integrated. In order to test the reliability of TSAIB_RS method, the paper takes the default risk of microcredit in China as the test target and compares the prediction results of various test methods. It concludes that TSAIB_RS method can significantly improve the prediction effects. [ABSTRACT FROM AUTHOR]