Abstract: Support vector regression machines based on the theory of statistical learning have a good generalization performance. However, in regression model if there are features that are not completely relative or even completely irrelative to the problem, the difference of features’ correlative degree to the problem becomes large and it may affect the performance of support vector regression machine. In order to solve the problem, support vector machine based on weighted feature is proposed in this paper. The results of simulation experiments show the feasibility and effectiveness of the method.

Key words: Support Vector Regression Machine, Weighted Feature, Machine learning

摘要： 基于统计学习理论的支持向量回归机有比较好的泛化能力，然而当样本含有与该问题不完全相关甚至完全无关的特征时，会使得各个特征对问题的相关程度差异很大，从而使得支持向量回归机的效果受到影响。为了解决这个问题，本文提出了一种基于特征加权的支持向量回归机。模拟的计算结果显示出此方法的有效性。

关键词: 支持向量回归机, 加权特征, 机器学习