Tourism demand forecasting by support vector regression and genetic algorithm

摘要：

Support vector regression optimized by genetic algorithm (G-SVR) is proposed to forecast tourism demand. Genetic algorithm (GA) is used to search for SVRs optimal parameters, and adopt the optimal parameters to construct the SVR models. This study examines the feasibility of SVR in tourism demand forecasting by comparing it with back-propagation neural networks (BPNN).The experimental results indicate that the proposed G-SVR model outperforms the BPNN based on mean absolute percentage error (MAPE).

关键词： support vector regression tourism demand neural networks auto-adaptive parameters

作者: Zhong-jian Cai Sheng Lu Xiao-bin Zhang

作者单位: School of Computer Science and Information Engineering Chongqing Technology and Business University Guangxi Special Equipment Supervision and Inspection Institute Nanning ,China

会议类型: 国际会议

会议名称: 2009 2nd IEEE International Conference on Computer Science and Information Technology(第二届计算机科学与信息技术国际会议 ICCSIT2009)

会议地点: 北京

会议语种:英文

页码: 2901-2903

在线出版日期: 2009-08-08（万方平台首次上网日期，不代表论文的发表时间）