Tracing Asian Giant Hornet: From a Data-Driven Perspective

Asian giant hornet is the largest wasp in the world. It has the characteristics of fast tracking speed and high toxicity, which poses a threat to human survival. Recently, footprints of the Asian giant hornet have been discovered on Vancouver Island and neighboring Washington State. Based on the types, seasonality, and expert evaluation information of witness reports, this article establishes multi-classification logistic regression and random forest model respectively, in which oversampling methods are used to solve the class imbalance problem. In addition, by evaluating the factor weights of the model, a type of timing strategy to update the model parameters is proposed, aiming to obtain higher generalization ability and robustness. The data-driven method can effectively reduce the misclassification of witness reports, thereby saving the labor and time required for expert labeling, which provides certain practical guiding significance.