论文部分内容阅读
针对兼类样本,提出一种类增量学习算法.利用超球支持向量机,对每类样本求得一个能包围该类尽可能多样本的最小超球,使各类样本之间通过超球隔开.增量学习时,对新增样本以及旧样本集中的支持向量和超球附近的非支持向量进行训练,使得算法在很小的空间代价下实现兼类样本类增量学习.分类过程中,根据待分类样本到各超球球心的距离判定其所属类别.实验结果表明,该算法具有较快的训练、分类速度和较高的分类精度.
Aiming at the two kinds of samples, a kind of incremental learning algorithm is proposed. Using the hypersphere support vector machine, a minimum hypersphere that can enclose as many samples as possible is obtained for each type of samples, In incremental learning, the new samples and the support vector in the old sample set and the non-support vector in the vicinity of the hypersphere are trained, so that the algorithm can realize the incremental learning of the sample type with a small space cost. , According to the distance between the samples to be tested and the center of each hypersphere, the classification of the class is determined.The experimental results show that the algorithm has fast training, classification speed and high classification accuracy.