A study of deep learning algorithm usage in predicting building loss ratio due to typhoons: the case of southern part of the Korean Peninsula

The goal of this study is to suggest an approach to predict building loss due to typhoons using a deep learning algorithm. Due to the influence of climate change, the frequency and severity of typhoons gradually increase and cause exponential destruction of building. Therefore, related industries and the government are focusing their efforts on research and model development to quantify precisely the damage caused by typhoons. However, advancement in the accuracy of prediction is still needed, and the introduction of new technology, obtained due to the fourth revolution, is necessary. Therefore, this study proposed a framework for developing a model based on a deep neural network (DNN) algorithm for predicting losses to buildings caused by typhoons. The developed DNN model was tested and verified by calculating mean absolute error (MAE), root mean square error (RMSE) and coefficient of determination ( R 2 ). In addition, to further verify the robustness of the model, the applicability of the framework proposed in this study was verified through comparative verification with the conventional multi-regression model. The results and framework of this study will contribute to the present understanding by suggesting a deep learning method to predict the loss of buildings due to typhoons. It will also provide management strategies to related workers such as insurance companies and facility managers.