Analysis on region grain security warning based on fuzzy least squares support vector machines

In order to enhance beforehand alarm precision on region grain security, due to the fuzzy speciality of beforehand alarm analysis on region grain security, a new fuzzy least squares support vector machines beforehand alarm model of region grain security based on chaos genetic algorithm is developed, in which the fuzzy membership function is set by using of clear sets to construct a fuzzy set and its parameters is optimized by using of chaos genetic algorithm. The application results revealed that beforehand alarm relative errors of the beforehand alarm model were less than 2.0%, and the connection of capability index parameters for region grain security can be expressed as: weight coefficient a1 of grain self- sufficiency rate > weight coefficient a2 of per capita share of grain > weight coefficient a3 of grain production levels> weight coefficient a4 of per capita arable land > weight coefficient a5 of fluctuation coefficient in grain production. Some theoretical evidence will be given for implementing beforehand alarm analysis on region grain security speedly and effectively by the research results.