中文摘要：

粒度支持向量机（Granular Support Vector Machine,GSVM）是以粒度计算理论和统计学习理论为基础的一种新的机器学习模型,它可以有效地克服传统支持向量机（Support Vector Machine,SVM）对于大规模数据集训练效率低下的问题,同时也可获得较好的泛化性能.文章针对原空间的GSVM模型进行了分析,提出了核空间的GSVM学习模型,在标准数据集上的实验说明了文中提出模型的有效性.

英文摘要：

Granular Support Vector Machine （GSVM） is a novel machine learning model based on granular computing theory and statistical learning theory. It can solve the low efficiency learning problem that existing in traditional Support Vector Machine（SVM）, and obtain the satisfactory generalization performance as well. This paper analyzes the GSVM models based on original space and proposes an GSVM model based on kernel space（KGSVM）. The experiments on benchmark datasets demonstrate the effectiveness of the Droposed approach.