|
|||||
|
|
| 基于最近边界向量的SVM增量学习算法 | |
| 引用本文： | 刘叶青,刘三阳,谷明涛.基于最近边界向量的SVM增量学习算法[J].数学的实践与认识,2011,41(2). |
| 作者姓名： | 刘叶青 刘三阳 谷明涛 |
| 作者单位： | 1. 河南科技大学,数学与统技学院,河南,洛阳,471003 2. 西安电子科技大学,理学院,陕西,西安,710071 3. 解放军96251部队,河南,洛阳,471003 |
| 基金项目： | 河南科技大学博士科研启动基金;河南科技大学青年基金(2008QN205) |
| 摘 要： | 为了减少求支持向量过程中二次规划的复杂度,利用训练样本集的几何信息,选出两类中离另一类最近的边界向量集合,它是样本中最有可能成为支持向量的一部分,用它代替原样本集进行训练.对新增样本,若存在违反KKT条件的样本,只对这部分新样本进行学习.同时找出原样本中可能转化为支持向量的非支持向量样本.基于分析结果,提出了一种新的基于最近边界向量的增量式支持向量机学习算法.对标准数据集的实验结果表明,算法是可行的,有效的. |
| 关 键 词： | 支持向量机 增量算法 数据挖掘 分类 |
Incremental Learning Algorithm of Support Vector Machine Based on Nearest Border Vectors |
|
| LIU Ye-qing,LIU Sanyang,GU Ming-tao.Incremental Learning Algorithm of Support Vector Machine Based on Nearest Border Vectors[J].Mathematics in Practice and Theory,2011,41(2). | |
| Authors: | LIU Ye-qing LIU Sanyang GU Ming-tao |
| Institution: | LIU Ye-qing~1,LIU San-yang~2,GU Ming-tao~3 (1.School of Mathematics and Statistics,Henan University of Science & Technology,Luoyang 471003,China) (2.Department of Applied Mathematics,Xidian University,Xi'an shaanxi 710071,China) (3.PLA Unit96251,China) |
| Abstract: | In order to reduce the time consumed in solving quadratic programming problems, a set of nearest border vectors were extracted from the training samples by using the geometric information in these samples.The original sample set was replaced by the obtained nearest border vector set in the process of training.The nearest border vector set is most likely to become the support vectors.For new samples,those was learned which do not satisfy Karush-Kuhn-Tucker(KKT) conditions.Besides support vectors,those was le... |
| Keywords: | support vector machine(SVM) incremental algorithm data mining classification |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |