|
|||||
|
|
| 单实例分类算法研究 | |
| 引用本文： | 潘志松,燕继坤,杨绪兵,缪志敏,陈斌.单实例分类算法研究[J].南京理工大学学报(自然科学版),2009,33(4). |
| 作者姓名： | 潘志松 燕继坤 杨绪兵 缪志敏 陈斌 |
| 作者单位： | 1. 解放军理工大学指挥自动化学院,江苏,南京,210007 2. 西南电子研究所,四川,成都,610041 3. 南京航空航天大学计算机科学与技术学院,江苏,南京,210016 |
| 摘 要： | 针对不平衡分类问题的极端情况,即用于训练的样本极少甚至只有一个实例,该文提出了一种单实例分类算法,这种方法使用球面作为分类面,在目标类的单实例在球内和反类尽量位于球面外的约束条件下,最大化该分类球面的半径,该方法能够有效地处理线性可分的数据分布.当输入样本分布结构呈高度非线性时,该算法通过核映射将低维输入空间中的非线性可分问题变换为高维特征空间中可能的线性可分问题,并以内积形式刻画,最终在特征空间上通过核技巧获得原问题的解决.通过对标准数据集和实际数据集的实验,验证了单实例分类算法在处理数据不平衡问题上的有效性. |
| 关 键 词： | 单实例 核方法 分类 支持向量 |
Classification Algorithm Based on Single Sample |
|
| PAN Zhi-song,YAN Ji-kun,YANG Xu-bing,MIAO Zhi-min,CHEN Bin.Classification Algorithm Based on Single Sample[J].Journal of Nanjing University of Science and Technology(Nature Science),2009,33(4). | |
| Authors: | PAN Zhi-song YAN Ji-kun YANG Xu-bing MIAO Zhi-min CHEN Bin |
| Institution: | 1.Institute of Command Automation;PLA University of Science and Technology;Nanjing 210007;China;2.The West-South Electronics Institute;Chengdu 610041;3.Department of ComputerScience and Engineering;Nanjing University of Aeronautics & Astronautics;Nanjing 210016;China |
| Abstract: | In order to solve the extreme situation that only a few target examples or only one can be used in training the classification,a single sample classification algorithm is presented here.Spherical surfaces are applied as classified hypersphere,and the largest radius can be obtained enclosing the single sample under the restriction that all outliers are outside the hypersphere.It fails when the distribution of input patterns is complex.The classifier applies kernel means,performing a nonlinear data transforma... |
| Keywords: | single samples kernel means classification support vectors |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |