肺结核病与气象因素关系的BP神经网络模型研究

目的 应用BP人工神经网络模型探讨气象因素对肺结核病发病影响,同时建立肺结核病与气象因素关系的BP神经网络模型.方法 利用Matlab 6.5的Statistics Neural Network软件对气象因素与肺结核病关系的BP人工神经网络模型进行构建、训练与模拟.结果 经过数据训练得出理想网络模型,肺结核病发病回代误差均方、平均误差率和R2分别为0.00713、0.82和0.9081,说明所得人工神经网络模型效果理想.通过对自变量对输出量贡献量分析表明,平均蒸发量对肺结核发病影响最大,平均气压亦有一定影响.结论 肺结核与气象因素关系的BP人工神经网络模型效果良好,有助于进一步研究的价值.

The Model of Back-propagation Neural Network about the Relationship between Meterological Factors and Pulmonary Tuberculosis

Objective In order to study the relationship between meterological factors and pulmonary tuberculosis. Methods Back-propagation artifical neural model was used by Matlab 6.5 statistics neural network to built the model of the relationship between meterological factors and pulmonary tuberculosis. Results The Mean Squared Errors was 0.00713, The Mean Error Rate and R^2 was 0.82 and 0.9081, it shown that the BP-neural network was effect. At the same time it shown that the mean evaporation and mean pressure was correlation to the incidence of pulmonary tuberculosis as well. Conclusion Bp-Neural Network model has effect of fiting on the relationship between meterological factors and pulmonary tuberculosis.