Multimedia based risk forecasting model for frequent natural disasters

Agricultural catastrophe risk assessment is not only a practical problem which needs to be solved for the stable development of China’s agriculture under the overall background of global climate change, but also a scientific issue which urgently needs to be solved for the discipline of agricultural risk assessment. At present, the academic circle mainly studies the theories and methods of agricultural catastrophe risk assessment from different perspectives based on risk factors, risk loss as well as risk mechanism, but fails to provide a clear answer to the loss degree of agricultural production caused by extreme weather events as well as the probability distribution of agricultural catastrophe loss. In view of this problem, and on the basis of modern risk analysis and assessment theory, a basic framework for agricultural catastrophe risk assessment was constructed in this thesis, and the parallel random tree algorithm was adopted to construct an agricultural economic risk prediction model in the environment of frequent natural disasters, which could be effectively applied to agricultural economic risk prevention and control.