四川农业大学学报 ›› 2008, Vol. 26 ›› Issue (03): 290-292.doi: 10.16036/j.issn.1000-2650.2008.03.020

• 研究简报 • 上一篇    下一篇

支持向量回归机在农业供应链预测中的应用

陈冬冬1,2, 杨春2   

  1. 1. 西南交通大学 物流学院, 四川 成都 610031;
    2. 四川农业大学 经济管理学院, 四川 雅安 625014
  • 收稿日期:2008-04-30 出版日期:2008-09-30 发布日期:2017-03-03
  • 作者简介:陈冬冬(1974~),男,四川荥经人,西南交通大学物流学院物流工程博士研究生,四川农业大学经济管理学院讲师。研究方向:物流工程、供应链管理、数量经济。

Applification of Support Vector Regression for Forecasting Agricultral Supply Chain

CHEN Dong-dong1,2, YANG Chun2   

  1. 1. College of Logistics, Southwest Jiaotong University, Chengdu 610031, Sichuan, China;
    2. College of Economics and Management, Sichuan Agricultural University, Yaan 625014, Sichuan, China
  • Received:2008-04-30 Online:2008-09-30 Published:2017-03-03

摘要: 为了提农业供应链预测的能力,应用基于结构风险最小化准则的标准支持向量回归机方法来研究供应链预测问题。在选择适当的参数和核函数的基础上,通过对实例研究,对时间序列数据进行预测,并与人工神经网络方法进行对比,发现该方法能获得最小的训练相对误差和测试相对误差。结果表明,支持向量回归机是研究农业供应链预测的有效方法。

关键词: 支持向量回归机, 农业供应链, 预测模型, SVM

Abstract: To improve the forecasting ability of agricultural supply chain,the method of support vector regression (SVR) can be applied to study the problem of agricultural supply chain,based on the principle of structural risk minimization.On the basis of choosing proper parameters and kernel functions,this method can make minimum relative errors in training and testing after making a study of examples,forecasting the time series data,and making a comparison with the method of Artificial Neural Network (ANN).The result shows that SVR is a useful way to study the forecasting of agricultural supply chain.

Key words: support vector regression, agricultural supply chain, forecasting models, SVM

中图分类号: 

  • F062.9