| 一种新的支持向量机增量学习算法 |
| |
| 引用本文： | 滕月阳,唐焕文,张海霞. 一种新的支持向量机增量学习算法[J]. 计算机工程与应用, 2004, 40(36): 77-80 |
| |
| 作者姓名： | 滕月阳 唐焕文 张海霞 |
| |
| 作者单位： | 大连理工大学应用数学系,大连,116024;大连理工大学应用数学系,大连,116024;大连理工大学应用数学系,大连,116024 |
| |
| 基金项目： | 国家自然科学基金重大研究计划资助项目(编号:90103033) |
| |
| 摘 要： | 支持向量机已经成为处理大规模高维数据的一种有效方法。然而处理大规模数据需要的时间和空间代价很高,增量学习可以解决这个问题。该文分析了支持向量的性质和增量学习的过程,提出了一种新的增量学习算法,舍弃了对最终分类无用的样本,在保证测试精度的同时减少了训练时间。最后的数值实验和应用实例说明:算法是可行的、有效的。
|
| 关 键 词： | 支持向量机 机器学习 增量学习 |
| 文章编号： | 1002-8331-(2004)36-0077-04 |
A New Algorithm to Incremental Learning with Support Vector Machine |
| |
| Teng Yueyang Tang Huanwen Zhang Haixia. A New Algorithm to Incremental Learning with Support Vector Machine[J]. Computer Engineering and Applications, 2004, 40(36): 77-80 |
| |
| Authors: | Teng Yueyang Tang Huanwen Zhang Haixia |
| |
| Abstract: | Support Vector Machine has become a popular tool for learning with large amounts of high dimensional data.However it's very costly to process these data in terms of time and memory consumption,then Incremental Learning may solve the problem.By analyzing the characteristics of support vectors and the processing of incremental learning,this paper presents a novel new algorithm of incremental learning.It discards useless samples and keeps the testing accuracy;meanwhile,training time is reduced.Numerical and applicable results illustrate that the technique is feasible and effective in the end. |
| |
| Keywords: | support vector machine machine learning incremental learning |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
