摘要
根据统计学习理论,间隔大小是反映泛化能力的一个很重要的方面.受一类支持向量机(SVM)的启发,提出的双边界SVM能分别用2个边界对2类问题分类.它能在保证分类正确的同时保证分类间隔的最大化,理论上分别从推广性能和不平衡类分布2方面证明了其优越性.标准数据集上的实验表明,双边界SVM得到的分类间隔要大于SVM,泛化性有了显著提高;另外,不平衡数据集上分析得到它对少数类识别率有明显提升.真实入侵数据测试结果表明,双边界SVM算法比边界样本选择算法的检测率高出2%以上.
Based on the statistical learning theory(SLT),the margin scale reflects the generalization capability to a great extent.Inspired by one-class support vector machine(SVM),double-margin SVM is put forward to classify two classes by two margins separately.Instances can be classified correctly as well as margin maximization,and its superiority is theoretical proved by both generalization performance and imbalanced class distribution.Experiment on benchmark data sets shows that classification margin obtained by double-margin SVM is larger than SVM,improving the generalization apparently,and analysis on imbalanced data shows that it has a higher recognition ratio.Finally real intrusion detection data shows that the detection precision is increased by 2% against boundary samples selection method.
出处
《北京邮电大学学报》
EI
CAS
CSCD
北大核心
2010年第2期20-23,共4页
Journal of Beijing University of Posts and Telecommunications
基金
国家高技术研究发展计划项目(2008AA01Z136)
关键词
分类间隔
泛化性能
双边界支持向量机
classification margin
generalization capability
double-margin support vector machine