论文部分内容阅读
类间间隔和类内聚类性是影响分类器分类性能的两种重要因素．基于模糊支持向量机和总间隔思想，提出一种基于总间隔的模糊v-相对间隔机（TMF-vRMM），本方法本质上是传统相对间隔机（RMM）的扩展，但可取得比RMM更好的分类性能．TMF-vRMM通过使用差异成本和引入总间隔和模糊隶属度，同时解决了不平衡训练样本问题和传统软间隔分类机RMM的过拟合问题，显著提升学习机的泛化能力．分别采用人造和实际数据集进行分类实验，结果显示TMF-vRMM具有优于相关方法的稳定分类性能．“,”The between-class margin and within-class cohesion are two important factors impacting on the performance of classifiers. This paper presents a novel classifier called total margin based fuzzy v-relative margin machine （TMF-vRMM） based on the idea of fuzzy support vector machine （FSVM） and total margin. Although it can be seen as a modified class of the classic RMM, TMF-vRMM has better theoretical classification performance than RMM. The proposed method solves not only the over-fitting problem resulted from outliers with the approaches of fuzzification of the penalty and total margin algorithm, but also the imbalanced datasets by using different cost algorithm, thus obtaining a lower generalization error bound. Experimental results obtained with synthetic and real datasets respectively show that the algorithm proposed in the paper is stable and superior to other related diagrams.