人工鱼群算法优化支持向量机的物流需求预测模型研究  

Logistics Demand Forecasting by Using Support Vector Machine Optimized by Artificial Fish Swarm Algorithm

作　　者：陈海英 张萍[2] 柳合龙[2] 

机构地区：[1]武汉设计工程学院,湖北武汉430205 [2]信阳师范学院数学与信息科学学院,河南信阳464000

出　　处：《数学的实践与认识》2016年第2期69-75,共7页Mathematics in Practice and Theory

基　　金：2012年湖北省高校省级教学研究项目(2012458);2014年华中农业大学楚天学院校级科研项目(K201405);河南省基础与前沿技术研究计划项目(142300410199;132400410784);2015年信阳师范学院青年基金项目

摘　　要：物流需求受多种因素的作用,具有时变性和混沌性,针对当前支持向量机的参数优化难题,提出一种改进人工鱼群算浅优化支持向量机的物流需求预测模型.首先对原始物流需求数据进行混沌分析,挖掘出隐藏其中的物流需求变化规律,然后采用支持向量机对物流需求数据进行非线性建模,并采用人工鱼群算法搜索支持向量机的参数,最后利用某地区物流数据与当前经典模型进行性能对比测试.结果表明,模型预测精度.更高,更加客观地反映了物流需求变化特性.Logistics demand is influenced by many factors, which is time variable and chaotic. In this study, a logistics demand forecasting model is proposed to solve the parameter optimiza- tion problem of support vector machine, which is optimized by artificial fish swarm algorithm. First, the chaos analysis of the original logistics demand is carried out to find the variation law. Then the support vector machine is used to model the nonlinear logistics demand data, and the artificial fish swarm algorithm is used to search the parameters of support vector machine. Its performance is tested by using many logistics data compared with the classical model. Results show that the proposed model has higher accuracy, which reflects the variation of logistics demand objectively.

关 键 词：支持向量机 区域物流需求 参数优化 人工鱼群算法 

分 类 号：F259.2[经济管理—国民经济] TP183[自动化与计算机技术—控制理论与控制工程]