用于不平衡数据分类的模糊支持向量机算法

作为一种有效的机器学习技术,支持向量机已经被成功地应用于各个领域.然而当数据不平衡时,支持向量机会产生次优的分类模型;另一方面,支持向量机算法对数据集中的噪声点和野点非常敏感.为了克服以上不足,提出了一种新的用于不平衡数据分类的模糊支持向量机算法.该算法在设计样本的模糊隶属度函数时,不仅考虑训练样本到其类中心距离,而且考虑样本周围的紧密度.实验结果表明,所提模糊支持向量机算法可以有效地处理不平衡和噪声问题.

A fuzzy support vector machine algorithm for imbalanced data classification

As an effective machine learning technique, support vector machine (SVM) has been successfully applied to various fields. However, when it comes to imbalanced datasets, SVM produces suboptimal classification models. On the other hand, the SVM algorithm is very sensitive to noise and outliers present in the datasets. To overcome the disadvantages of imbalanced and noisy training datasets, a novel fuzzy SVM algorithm for imbalanced data classification is proposed. When designing the fuzzy membership function, the proposed algorithm takes into account not only the distance between the training sample and its class center, but also the tightness around the training sample. Experimental results show that the proposed fuzzy SVM algorithm can effectively handle the imbalanced and noisy problem.