Abstract: In this paper,a novel kernel-based fuzzy multiple spheres classification algorithm is proposed.In the training process,all training samples of each class are covered by the constructed multiple spheres,each of which encompasses as many samples with the same class and has the minimum volume,and a fuzzy membership function is defined to label the testing samples in the classification process.Moreover,an ensemble method based on the proposed classification algorithm is presented.Finally,experiments on four real datasets show that our approach is valid and has encouraging pattern classification performance.

Key words: pattern classification, kernel function, mountain function, fuzzy membership function, classification ensemble

摘要： 提出了一种基于核的模糊多球分类算法,该算法在训练阶段为每一个模式类构造多个最小球覆盖其所有的训练样本,并且在识别阶段算法利用一个模糊隶属函数来归类测试样本。此外,在提出的分类算法的基础上,还给出了它的集成方法。最后,采用了4个真实数据集进行实验,实验结果表明该文提出的算法具有较好的分类性能,是一种行之有效的分类算法。

关键词: 模式分类, 核函数, 山峰函数, 模糊隶属函数, 分类集成