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 ›› 2008, Vol. 34 ›› Issue (19): 86-88. doi: 10.3969/j.issn.1000-3428.2008.19.030

• 网络与通信 • 上一篇    下一篇

网络信息过滤系统的过载处理方法研究

余智华1,2   

  1. (1. 中国科学院计算技术研究所智能软件部,北京 100080;2. 中国科学院研究生院,北京 100049)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-10-05 发布日期:2008-10-05

Research on Anti-overload Methods in Network Information Filtering System

YU Zhi-hua1,2   

  1. (1. Intelligent Software Division, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100080; 2. Graduate University of Chinese Academy of Sciences, Beijing 100049)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-10-05 Published:2008-10-05

摘要: 针对网络信息过滤系统的过载问题,通过牺牲少量准确率和召回率来提高系统检测能力,并使用采样和规则泛化等近似处理技术保持系统在过载情况下的总体过滤能力。实验结果表明,在准确率或召回率仅降低10%~15%的情况下,能使系统可承受负载提高3倍,性能提升效果显著。

关键词: 信息过滤, 过载处理, 采样, 泛化规则

Abstract: Due to the overload condition for the network information filtering system, this paper proposes the strategy that improves filtering speed paying a little loss of precision and recall, uses approximate methods such as text sampling and pattern generalizing. It can retain most of the filtering capability in overload condition. The experimental results show that the filtering capacity can be improved about three times while losing only less than 15% of precision or recall and has a remarkable performance promotion.

Key words: information filtering, anti-overload, sampling, generalized rules

中图分类号: