| 基于径向基函数(RBF)神经网络的红鳍东方鲀体质量预测 |
| |
| 引用本文： | 王新安,马爱军,赵艳飞,岳亮,孙建华,孟雪松,刘圣聪.基于径向基函数(RBF)神经网络的红鳍东方鲀体质量预测[J].水产学报,2015,39(12):1799-1806. |
| |
| 作者姓名： | 王新安 马爱军 赵艳飞 岳亮 孙建华 孟雪松 刘圣聪 |
| |
| 作者单位： | 中国水产科学研究院黄海水产研究所,中国水产科学研究院黄海水产研究所,中国水产科学研究院黄海水产研究所，荣成市渔业技术推广站,中国水产科学研究院黄海水产研究所,大连天正实业有限公司,大连天正实业有限公司,大连天正实业有限公司 |
| |
| 基金项目： | 国家现代农业技术体系资助项目(CARS-50-G01);中央级公益性科研院所基本科研业务费专项(20603022012005);大连金州新区科技计划(2012-B1-012) |
| |
| 摘 要： | 基于表型性状预测红鳍东方鲀体重时,由于不同表型性状间的自相关、部分性状和体重之间的非线性关系以及线性回归方法自变量间的共线性,结果导致预测误差过大。为了解决这一问题,根据人工神经网络(Artificial Neural Networks, ANN)建模原理,采用径向基函数(Radial Basis Function, RBF) 模型,利用72个红鳍东方鲀样品的生长数据, 通过最近邻聚类算法, 构建了基于RBF神经网络的红鳍东方鲀体重预测模型,并采用线性回归检验法对所构建模型的可信度进行检验。结果发现,基于RBF神经网络预测模型的拟合优度为0.992,接近于1,而线性回归模型的拟合度为0.949。这表明: 基于RBF神经网络方法构建的预测模型消除了线性分析中自变量的共线性问题, 比线性回归预测模型的拟合度提高4.53%,预测精度高于线性回归方法。基于RBF神经网络体重预测模型的构建,为利用表型性状精确评估红鳍东方鲀的体重提供了一种新的方法。
|
| 关 键 词： | 红鳍东方鲀 体重预测 径向基函数 神经网络 |
| 收稿时间： | 2015/3/23 0:00:00 |
| 修稿时间： | 9/5/2015 12:00:00 AM |
Prediction of Takifugu rubripes weight based on radial basis function neural network |
| |
| WANG Xin''an,MA Aijun,ZHAO Yanfei,YUE Liang,SUN Jianhu,MENG Xuesong and LIU Shengcong.Prediction of Takifugu rubripes weight based on radial basis function neural network[J].Journal of Fisheries of China,2015,39(12):1799-1806. |
| |
| Authors: | WANG Xin'an MA Aijun ZHAO Yanfei YUE Liang SUN Jianhu MENG Xuesong and LIU Shengcong |
| |
| Institution: | Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences,Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences,1. Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences; Key Laboratory of Sustainable Development of Marine Fisheries, Ministry of Agriculture; Qingdao Key Laboratory for Marine Fish Breeding and Biotechnology, Qingdao, 266071, China; 2. Rongcheng Fisheries Technical Extension Station 264300, China,Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences,Dalian Tianzheng Industrial Co. Ltd., Dalian,Dalian Tianzheng Industrial Co. Ltd., Dalian,Dalian Tianzheng Industrial Co. Ltd., Dalian |
| |
| Abstract: | There existSlarge error due to self-correlation between different phenotypic traits, non-linear relationship between some traits and body weight and the collinearity among independent variables, when the linear regression model was used to predict Takifugu rubripes weight. AsSa solution, a Takifugu rubripes weight prediction RBF neural network model, according to Artificial Neural Networks theory and Radial Basis Function model,was constructed with the phenotypic traits of 72 Takifugu rubripes based on the nearest neighbor clustering algorithm, and the credibility of the neural network modelSconstructedS was test by linear regression techniques. The regression analysis demonstrated that the R2 of RBF neural network prediction model for Takifugu rubripes weight was 0.992 and higher than that of linear regression model. The results suggested that the RBF neural network techniques was an effective method to construct the prediction model of Takifugu rubripes. The collinearity of the independent variables, in RBF neural network analysis, was eliminated and itShas higher accuracy than linear regression prediction model. Weight prediction model based on radical basis function neural network provideSaSnew method for accurately prediction of Takifugu rubripes weight. |
| |
| Keywords: | Takifugu rubripes Weight prediction Radical basis function Neural network |
|
| 点击此处可从《水产学报》浏览原始摘要信息 |
| 点击此处可从《水产学报》下载免费的PDF全文 |
|
