查询结果:   霍静,王永明,顾君忠.感染性腹泻周发病例数的PCA-SVM回归预测研究[J].计算机应用与软件,2016,33(2):51 - 54.
中文标题
感染性腹泻周发病例数的PCA-SVM回归预测研究
发表栏目
应用技术与研究
摘要点击数
821
英文标题
RESEARCH ON PCA-SVM REGRESSIVE PREDICTION OF WEEKLY CASES OF INFECTIOUS DIARRHEA
作 者
霍静 王永明 顾君忠 Huo Jing Wang Yongming Gu Junzhong
作者单位
天水师范学院电子信息与电器工程学院 甘肃 天水 741001 华东师范大学计算机应用研究所 上海 200062    
英文单位
School of Electronic Information and Electronical Engineering,Tianshui Normal University,Tianshui 741001,Gansu,China Institute of Computer Applications,East China Normal University,Shanghai 200062,China    
关键词
PCA SVM回归 感染性腹泻 气象资料
Keywords
PCA SVM regression Infectious diarrhea Meteorological data
基金项目
上海市国际科技合作基金项目(134307 10100);甘肃省科技计划资助项目(1506RJZE115);甘肃省高等学校科研项目(2015B-104)
作者资料
霍静,讲师,主研领域:数据挖掘,信息检索。王永明,博士。顾君忠,教授。 。
文章摘要
提出一个使用PCA-SVM进行感染性腹泻周发病例数回归预测方法, 有效避免了BP神经网络模型存在局部极值、多重共线性的问题。以上海市2005年至2008年感染性腹泻周发病例数为样本,建立PCA-SVM回归模型。首先用PCA从统计气象因子中提取气象主成分因子, 去除预报因子多重共线性,得到最终模型的解释变量, 其次采用SVM方法构建上海市感染性腹泻周发病例数预测模型。为了说明该模型有更佳的预测效果,与BP神经网络模型比较拟合及预测结果。数据结果显示PCA-SVM回归模型预测的平均相对误差MAPE、均方误差平方根RMSE(数值分别为0.2694,33.113)均小于BP神经网络(数值分别为0.3745,49.909),而决定系数R2 (数值为0.9089)较BP神经网络(数值为0.8590)更趋近于1。证明PCA-SVM回归模型在感染性腹泻周发病例数预测中具有较高的预测精度和较强的泛化能力,模型对于感染性腹泻周发病例数的预测可靠,对于向公众发布腹泻预报有更好的实用价值。
Abstract
We proposed a regressive prediction method for the weekly cases number of infectious diarrhea using PCA-SVM, which effectively avoids some defects of the BP neural network model like local extremum, multicollinearity. With the weekly cases of infectious diarrhea in Shanghai from the year 2005 to 2008 being the samples, we built the PCA-SVM regressive model. First, we employed PCA to extract meteorological main principal factors from the statistical meteorological factors and removed the multicollinearity from the predictive factors, derived the explanatory variable of the final model. Secondly, we used SVM regression to build the predictive model for weekly cases number of infectious diarrhea in Shanghai. To illustrate the better prediction effect of the model, we compared it with BP neural network model in terms of fitting and prediction results. Numerical results showed that the MAPE and RMSE (0.2694 and 33.113 respectively) predicted by PCA-SVM regression model were all less than those of BP neural network model (0.3745 and 49.909 respectively). Meanwhile, its determination parameter R2 (0.9089) was further approaching 1 than that of BP neural network (0.8590). As a result, it is demonstrated in this paper that the PCA-SVM regressive model has higher prediction accuracy and stronger generalisation capability in predicting weekly cases number of infectious diarrhea, the prediction of the model is reliable on the weekly cases number of the disease, and has better practical value in publicising the diarrhea prediction.
下载PDF全文