Factor selection and regression for forecasting relief food demand

Predicting relief food demand effectively and accurately after nature disasters is a key to maintain the life of victims. Currently, the main methods for forecasting relief food are based on the experts and the predication results are influenced by the experiences of the experts. So how to predicate the relief food demand based on the obtained nature disaster cases is a very important problem. In this paper we present a novel method to predicate the relief food demand using support vector machine. To select the factors which have influence on relief food demand, recursive feature elimination algorithm is adopted. The experimental results on real disaster cases of Hubei province of China prove the performance of the proposed method.