多核支撑向量回归方法研究  

Research on Multiple-Kernel Support Vector Regression Method

作　　者：陈博 郑凯东[1] 王家华[1] CHEN Bo;ZHENG Kaidong;WANG Jiahua(School of Computer Science,Xi'an Shiyou University,Xi'an 710065)

机构地区：[1]西安石油大学计算机学院,西安710065

出　　处：《智能计算机与应用》2019年第1期188-191,共4页Intelligent Computer and Applications

摘　　要：近些年来,支撑向量回归方法在减少泛化误差方面表现出了出色的性能。然而,传统的支撑向量机或者支撑向量回归方法是基于单个核函数的,在高维空间中解决非线性问题。但随着应用领域不断扩展,在一些复杂情形下,由单个核函数构成的支撑向量回归方法并不能满足数据异构、输入空间维度过高等实际问题。针对此问题,人们在单核学习的基础上提出了多核学习,即将多个核函数进行线性组合,以此来提高模型精度,并逐渐成为当下机器学习领域研究的热点。文章综述了支撑向量回归算法与多核学习算法理论知识,并分析了各自的特点以及应用领域。总结了多核支撑向量回归方法下一步的研究趋势。In recent years,the Support Vector Regression method has shown excellent performance in reducing generalization errors.However,traditional Support Vector Machines or Support Vector Regression methods are based on a single kernel function to solve nonlinear problems in high-dimensional space.However,with the continuous expansion of the application field,in some complicated situations,the Support Vector Regression method consisting of a single kernel function can not meet the practical problems of data heterogeneity and input space dimension is too high.Aiming at this problem,multiple kernel learning has been put forward on the basis of single-core learning,which is to linearly combine multiple kernel functions to improve the accuracy of the model,and gradually become a hot research topic in the field of machine learning.This paper reviews the theory of Support Vector Regression algorithm and multiple kernel learning algorithm,and analyzes their respective characteristics and application fields.At the end of the article,the research trends of the multiple Kernel Support Vector Regression method are summarized.

关 键 词：支撑向量机 支持向量回归 多核学习 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]