论文部分内容阅读
[目的]探讨反向传播(back propagation,BP)神经网络模型与径向基函数(radial basis function,RBF)神经网络模型在煤工尘肺发病工龄预测中预测性能的优劣。[方法]采用SPSS 19.0中的BP和RBF神经网络模型对研究数据进行预测分析。采用均方根误差、平均相对误差和平均绝对误差对两种模型的预测结果进行比较分析,从而得到比较各模型预测性能的目的。[结果]由BP神经网络模型进行预测可以得出发病工龄预测值和真实值的散点分布图大致位于从原点起始的45度线上,符合理想状态下值的分布情况,而RBF神经网络模型的分布则较为混乱。RBF和BP神经网络模型真实值与预测值之间的差异均无统计学意义,t值分别为0.231、0.530,P值分别为0.817、0.596。RBF和BP神经网络模型的均方根误差分别为3.51、1.89;平均相对误差分别为12%、6%;平均绝对误差分别为2.76、1.42。[结论]实证表明,在煤工尘肺发病工龄的预测中,BP神经网络模型的预测性能优于RBF神经网络模型,有较高的拟合和预测精度。
[Objective] The purpose of this study was to discuss the pros and cons of predictive performance in predicting the age of coal workers’ pneumoconiosis with back propagation (BP) neural network model and radial basis function (RBF) neural network model. [Methods] The BP and RBF neural network model in SPSS 19.0 was used to predict the data. The root mean square error, average relative error and average absolute error are used to compare the prediction results of the two models, and the purpose of comparing the prediction performance of each model is obtained. [Result] The result of BP neural network model predicts that the scatter plot of predictive value and real value of onset age is roughly located on the 45 degree line from the origin, which is consistent with the distribution of ideal value. RBF neural network The distribution of the model is more confusing. There was no significant difference between the real value and the predicted value of RBF and BP neural network model, the t values were 0.231 and 0.530 respectively, and the P values were 0.817 and 0.596 respectively. The root mean square errors of RBF and BP neural network models were 3.51 and 1.89, respectively; the average relative errors were 12% and 6% respectively; the average absolute errors were 2.76 and 1.42 respectively. [Conclusion] Empirical results show that the prediction performance of BP neural network model is better than that of RBF neural network model in predicting the length of service life of coal worker ’s pneumoconiosis with high fitting and prediction accuracy.