基于PCA改进的LS-SVM入境旅游客流量预测模型  

The Forecast Model of Tourist Flow of Entering based on LS-SVM Improved by Principal Component Analysis

作　　者：张朝元[1] 杨泽恒[1] 王彭德[1] 陈丽[2] 

机构地区：[1]大理学院数学与计算机学院,云南大理671003 [2]大理学院工程学院,云南大理671003

出　　处：《科技通报》2012年第7期75-79,共5页Bulletin of Science and Technology

基　　金：云南省教育厅科学研究基金项目(2010C140)

摘　　要：影响入境旅游客流量的众多因素加大了预测模型输入变量的复杂化,限制了模型的运行速度和预测精确。首先,利用主成分分析对影响入境旅游客流量的众多指标进行综合分析得到主成分,然后建立以主成分为输入变量以入境旅游客流量为输出变量的最小二乘支持向量机预测模型。通过实例验证和比较,展示了基于主成分分析改进的最小二乘支持向量机入境旅游客流量预测模型具有较好的预测效果和较高的推广价值。Many factors influencing tourist flow of entering increase the complexity of the input variables and limit speed and precision of prediction model. Firstly, the main components are gained through analyzing the impact indicators of tourist flow of entering using principal component analysis. Secondly, it is established the forecast model of Least Squares Support Vector Machine based on input variables of main components and output variable of tourist flow. Through example confirmation and comparison, it is showed that good forecast effect and high application value of the forecast model of tourist flow of entering based on Least Squares Support Vector Machine improved by principal component analysis.

关 键 词：入境旅游客流量 主成分分析 最小二乘支持向量机 预测模型 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] F590[自动化与计算机技术—计算机科学与技术]