Prediction of emergency supplies based on improved Case-based reasoning

Abstract

In order to overcome the problems of old cases on prediction accuracy in traditional case-based reasoning (CBR), an improved case-based reasoning (CBR) method for emergency materials forecasting is proposed. The method adapts to the dependence of emergency events on social development by weakening the weight of old cases. Firstly, the coefficient of variation method is used to calculate the weight coefficient between case attributes. Secondly, the consumption strategy is used to reduce the weight of the old cases, and then the similarity between the cases is calculated to retrieve the best source case matching the target case. Finally, the simulation results show that the prediction accuracy of the improved method is proved to be better than that of the traditional method. Therefore, this method has a certain reference value for the prediction of emergency rescue materials.