基于径向基函数神经网络的组合模型在煤工尘肺发病工龄预测中的应用

发布时间：2018-06-11 20:41

本文选题：径向基函数神经网络 + 多重线性回归模型　；参考：《郑州大学学报(医学版)》2014年06期

【摘要】：目的:研究径向基函数(RBF)神经网络与多重线性回归的组合模型在煤工尘肺发病工龄预测中的性能优劣。方法:采用RBF神经网络模型与多重线性回归模型对研究数据进行分析,对2模型进行加权拟合,采用均方根误差、均方误差、平均相对误差对模型的预测性能进行评价。结果:多重线性回归模型、RBF神经网络模型和组合模型真实值与预测值比较,差异均无统计学意义(t配对=1.552、0.231、0.155,P均0.05)。多重线性回归模型、RBF神经网络模型和组合模型的均方根误差分别为(1.63±0.11)、(2.45±0.19)和(0.59±0.07)(F=26.141,P0.001),均方误差分别为(2.656 9±0.241 2)、(5.986 7±0.380 4)和(0.348 3±0.065 3)(F=49.678,P0.001),平均相对误差分别为(7.15±0.82)%、(15.39±1.25)%和(3.68±0.59)%(F=35.282,P0.001)。结论:在煤工尘肺发病工龄的预测中,组合模型预测性能优于单一模型。
[Abstract]:Objective: To study the performance of the combination model of the radial basis function (RBF) neural network and multiple linear regression in the prediction of the age of coal miner's pneumoconiosis. Method: the RBF neural network model and multiple linear regression model are used to analyze the research data, and the 2 model is weighted fitting, and the mean square error, mean square error and average relative error are used. The prediction performance of the model was evaluated by error. Results: the multiple linear regression model, the RBF neural network model and the combination model compared with the predicted value, the difference was not statistically significant (t pairing =1.552,0.231,0.155, P 0.05). Multiple linear regression model, RBF neural network model and the combination model of the root mean square error was (1.63 + 0). (.11), (2.45 + 0.19) and (0.59 + 0.07) (F=26.141, P0.001), the mean square error is (2.6569 + 0.2412), (5.9867 + 0.3804) and (0.3483 + 0.0653) (0.3483 + 0.0653) (F=49.678, P0.001), the average relative error is (7.15 + 0.82)%, (15.39 + 1.25)% and (F=35.282, P0.001). Conclusion: in the prediction of the age of coal workers' pneumoconiosis, the combined model is predefined. The test performance is better than the single model.
【作者单位】：河北省煤矿卫生与安全实验室;
【基金】：河北省科技支撑项目11276911D 河北省卫生厅医学重点项目20120146 唐山市科技支撑项目11150205A-3
【分类号】：R135.2

【参考文献】