摘要
对现有的模糊支持向量机进行分析,提出一种改进的模糊支持向量机算法——双隶属度模糊支持向量机法(DM-FSVM)。在传统的模糊支持向量机模型中,每一个训练样本的隶属函数中只有一个隶属度,而DM-FSVM中每一个训练样本拥有两个隶属度。它既能保持传统模糊支持向量机的优点,又能充分利用有限样本,增加其分类推广能力。实验表明该算法较好地提高了分类精度。
Based on traditional fuzzy support vector machine (FSVM), a new fuzzy support vector machine, dual membership fuzzy support vector machine (DM-FSVM), was presented. There is only one membership in traditional support vector machine (SVM) model; however, there are two memberships in DM-FSVM. This method not only keeps the advantages of traditional FSVM, but also makes full use of limited data and improves the classification efficiencies. Experiments show that DM-FSVM improves the classification accuracy.
出处
《计算机应用》
CSCD
北大核心
2007年第11期2821-2823,共3页
journal of Computer Applications
基金
江西省自然科学基金资助项目(0411046)
江西省高性能计算技术重点实验室资助基金项目(JXHC220052003)
江西省科技厅工业攻关项目(赣财教[2005]132号)
关键词
支持向量机
模糊支持向量机
隶属度
双隶属度
Support Vector Machine (SVM)
Fuzzy Support Vector Machine (FSVM)
membership
dual membership