基于FOA-RBF神经网络的外贸出口预测

A Forecast of Export Trades Based on the FOA-RBF Neural Network

应用果蝇优化算法对径向基神经网络扩展参数的优化方法进行研究,给出了一种以标准误差计算公式为味道判定函数,以此确定最优的径向基函数的扩展参数值的方法,并建立了相应的预测模型.应用该预测模型对黑龙江省外贸出口额进行预测,结果表明:预测模型的预测精度优于径向基神经网络,从而证明了方法的有效性.

This paper studied the optimized method of RBF neural network expansion parameter by the fruit fly optimization algorithm, which has given a method of take the standard error formula as the flavor decision function, and to determined the most superior method, and has established the corresponding forecast model. The finally indicated of Heilongjiang Province export trades amount forecasting with this forecast model is that this forecast model forecast precision is better than the RBF neural network, thus has proven this method validity.