论文部分内容阅读
目的为提供更加准确的预测结果合理调配卫生资源,将基于PSO的BP神经网络应用到铜陵市腮腺炎发病率预测中。方法根据铜陵市2006-2011年的腮腺炎发病率及其相关因素,利用基于PSO的BP神经网络建立腮腺炎发病率预测模型,并对模型的有效性进行验证。结果预测结果显示,PSO-BP神经网络对2012年1-6月铜陵市腮腺炎发病率的预测平均误差为2.57%,预测精度明显高于传统的BP神经网络。结论 PSO-BP神经网络模型能够较好地克服传统BP神经网络易陷入局部极值、收敛速度慢等缺点,为腮腺炎发病率的预测提供了一种有效的方法。
Objective To provide more accurate prediction results to rationally allocate health resources and apply PSO-based BP neural network to predict the incidence of mumps in Tongling City. Methods According to the morbidity and related factors of mumps from 2006 to 2011 in Tongling City, the prediction model of the incidence of mumps was established by BP neural network based on PSO, and the validity of the model was verified. Results The prediction results showed that the average prediction error of PSO-BP neural network for the incidence of mumps in Tongling from January to June in 2012 was 2.57%, and the prediction accuracy was significantly higher than that of the traditional BP neural network. Conclusion PSO-BP neural network model can better overcome the traditional BP neural network easily fall into the local extremes, slow convergence and other shortcomings, mumps for the prediction of the incidence provides an effective method.