基于最小二乘支持向量机的特征增量学习算法

An Incremental Feature Learning Algorithm for the Least Square Support Vector Machine

本文针对传统的增量学习算法无法处理后采集到的样本中含有新增特征的问题,设计适应样本特征维数增加的训练算法。在基于最小二乘支持向量机的基础上,提出了特征增量学习算法。该算法充分利用先前训练得到的分类器的结构参数,仅对新增特征采用最小二乘支持向量机进行学习。实验结果表明,该算法能够在保证分类精度的同时,有效地提高训练速度并降低存储空间。

In order to tackle with the incremental learning problems with new features, an incremental feature learning algorithm for the least square support vector machine is proposed in this paper. In this algorithm, using historic structural parameters trained from the already existing features, the algorithm only trains the new features with the least square support vector machine Experiments show that this algorithm has two outstanding properties. First, different kernel functions can be used for the already existing features and the new features according to the distribution of samples. Second, the training time and the memory space can be reduced. Some UCI datasets are used to demonstrate the less training time or the better performance of this algorithm than the standard least square support vector machine.