摘要
针对现有单类分类器对目标数据先验信息考虑的不足,在结构单类支持向量机(structured one-class supportvector machine,SOCSVM)中嵌入局部密度信息,提出局部密度嵌入的结构单类支持向量机(SOCSVM with local den-sity embedding ldSOCSVM)。借助K近邻(K-nearest neighbor,KNN)揭示目标数据局部密度,并进一步诱导出权重因子作用于样本点。该算法充分利用目标数据的全局信息及局部密度信息,从而提高分类器的泛化能力。UCI数据集上的实验结果验证了ldSOCSVM的有效性。
To improve the generalization ability of one-class classifier, more prior knowledge were taken into account on the existed models. A new structured one-class support vector machine with local density embedding (ldSOCSVM) was proposed, which could embed local information of target data into the structured one-class support vector machine (SOCSVM). By means of K-nearest neighbor, the weighted factor was extracted and applied to the corresponding sam- ples by fully utilizing local information with the global ones inherited from SOCSVM, the ldSOCSVM improved the generalization ability. Experimental results on UCI datasets showed that the proposed classifier could achieve better gen- eralization capability compared with related algorithms.
出处
《山东大学学报(工学版)》
CAS
北大核心
2012年第4期13-18,共6页
Journal of Shandong University(Engineering Science)
基金
国家自然科学基金资助项目(61170152)
关键词
单类分类器
先验信息
结构单类支持向量机
局部密度
权重因子
one-class classifier
prior knowledge
structured one-class support vector machine
local density
weighted factor