Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (27): 157-158.DOI: 10.3778/j.issn.1002-8331.2008.27.050

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Incremental learning algorithm for one-class document classification

DAI Hong1,ZHU Ming2,LIU Shou-qun2   

  1. 1.Department of Electronic Engineering and Information Science,University of Science & Technology of China,Hefei 230027,China
    2.Department of Automation,University of Science & Technology of China,Hefei 230027,China
  • Received:2007-11-09 Revised:2008-01-31 Online:2008-09-21 Published:2008-09-21
  • Contact: DAI Hong

支持增量学习的文本单类别分类算法

戴 洪1,朱 明2,刘守群2   

  1. 1.中国科学技术大学 电子信息工程系,合肥 200027
    2.中国科学技术大学 自动化系,合肥 200027
  • 通讯作者: 戴 洪

Abstract: In this paper,an incremental learning algorithm for one-class document classification is proposed.It also has the advantage of low computational load with the same level of performance.A prototype system is constructed to implement the algorithms,and the results of the tests are pretty good.

Key words: Naï, ve Bayesian, Support Vector Machine(SVM), one-class classification, text/Web classification

摘要: 目前的文本单类别分类算法在进行增量学习时需要进行大量的重复计算,提出了一种新的用于文本的单类别分类算法,在不降低分类效果的同时,有效地减少了加入新样本学习时所需的计算量,从而比较适合于需要进行增量学习的情况。该方法已进行了测试实验,获得了较好的实验结果。

关键词: 简单贝叶斯, 支持向量机, 单类别分类, 文本/网页分类