Description: web分类系统 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 classification system for Web document classification is a Web mining, one of the most basic skills, but to construct a classifier classified according to interest, need to do a lot of pre-processing work to collect the training sample of plus or minus. However, the collection of negative cases is very difficult. This paper presents a no negative cases only the positive cases of the learning model. The model is mainly repeat the SVM. Experiments show that the learning model for Web document categorization classification accuracy and speed are very good Platform: |
Size: 1232896 |
Author:huihui |
Hits:
Search - WebCategorization