作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2009, Vol. 35 ›› Issue (22): 212-215. doi: 10.3969/j.issn.1000-3428.2009.22.073

• 人工智能及识别技术 • 上一篇    下一篇

改进的在线支持向量机训练算法

潘以桢,胡越明   

  1. (上海交通大学计算机科学与工程系,上海 200240)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-11-20 发布日期:2009-11-20

Improved Online Training Algorithm of Support Vector Machine

PAN Yi-zhen, HU Yue-ming   

  1. (Department of Computer Science and Engineering, Shanghai Jiaotong University, Shanghai 200240)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-11-20 Published:2009-11-20

摘要: 传统支持向量机基于批量训练方法,无法适应环境污染预测中的海量数据与实时性要求。在分析研究一种典型的在线支持向量机回归算法[4]的基础上,指出原算法在训练过程中存在样本重复移动问题,导致模型训练速度下降。提出一种改进算法,消除重复移动问题。实验结果表明,该改进在线支持向量机算法建模精度高,训练速度较原算法有显著提高。

关键词: 污染预测, 支持向量机, 在线学习, 增量式学习

Abstract: Traditional Support Vector Machine(SVM), which based on batch training, can’t satisfy the real-time requirement of environmental pollution prediction with large scale data. With the analysis of a typical kind of online support vector regression algorithm, this paper indicates that repeated sample move exists in the training process would lead to decrease the training speed, and proposes an improved algorithm. Simulation and analysis results show that the proposed algorithm performs high modeling precision, and training speed is increased remarkably compared with the aforementioned algorithm.

Key words: pollution prediction, Support Vector Machine(SVM), online learning, incremental learning

中图分类号: