基于梯度提升决策树(GBDT)算法的南方洪涝灾害房屋倒损评估模型

A Model of House Damage Assessment for Southern China Based on Gradient Boosting Decision Tree (GBDT) Algorithm

结合影响洪涝灾害损失的10个指标,利用国家减灾中心灾情库中案例作训练样本,建立了基于梯度提升决策树(GBDT)算法的南方洪涝灾害房屋倒损评估模型,可用于洪涝灾害房屋倒损的快速评估。最后,利用2016年7月上旬南方洪涝灾害案例进行了模型验证,总体相对误差为17.9%,证明该模型具有较高的可靠性,能够较好反映洪涝损失总体情况。

In this research, a model based on Gradient Boosting Decision Tree(GBDT) algorithm for assessing house damage in Southern China is built. It can be used for house damage rapid assessment in flood disaster. This model is trained using flood samples from National Disaster Reduction Center of China(NDRCC), combining 10 indexes which might have effect on flood disaster loss. In the end, a flood case of Southern China in Early July 2016 is used to verify the model. The overall relative error is only 17.9%, which proves the model is of high reliability and reflects the overall flood disaster loss very well.