摘要
Web文档分类是Web挖掘中最基本的技术之一,而构造一个按照兴趣分类的分类器,需要做大量的预处理工作,来收集正负的训练样例。但负例的收集是非常困难的。文章提出了一个只有正例没有负例的学习模型。该模型主要是重复执行SVM。实验表明,该学习模型对于Web文档分类的分类精度和速度都是非常理想的。
Web page classification is one of the basic techniques for Web mining. However, to collecting positive and negative training examples, constructing a classifier for an interesting calss requires laborious preprocessing. But collecting negative examples is very diffcult. In this paper , a modle with positive examples, is brought forward without negative. The modle outperforms mainly SVM. The experiments show that this modle is very ideal for precision and speed of Web page classification.
出处
《微计算机应用》
2005年第4期432-435,共4页
Microcomputer Applications
