Multi-output LS-SVR Algorithm based on Extended Feature Space

摘要：

Support vector regression is traditionally used with only one output.Previous multi-output regression model is made up of multiple single-output regression models.Taking into account the correlations between multi-outputs,this paper presents a new method to construct multi-output model directly.The method extents the original feature space by vector virtualization.In the extended feature space,a single output least square support vector regression machine is designed to deal with multi-output problems in the original feature space.Experimental results show that this method presents good performance.

关键词： multi-output regression Is-svr extended feature sapce vector virtualization

作者: Wei Zhang Weidong Zhao

作者单位: College of Electronic and Information Engineering,Tongji University Shanghai,China

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer and Network Technology(ICCNT 2011)(2011第三届IEEE计算机与网络技术国际会议)

会议地点: 太原

会议语种:英文

页码: 237-240

在线出版日期: 2011-02-26（万方平台首次上网日期，不代表论文的发表时间）