| 基于n-gram及SVM的中文垃圾邮件过滤 |
| |
| 引用本文： | 夏成锋.基于n-gram及SVM的中文垃圾邮件过滤[J].广东广播电视大学学报,2008,17(1):100-103. |
| |
| 作者姓名： | 夏成锋 |
| |
| 作者单位： | 仲恺农业技术学院,广东广州,510225 |
| |
| 摘 要： | 在基于内容的垃圾邮件过滤方法中，特征表达和分类算法十分重要，本文应用n-gram方法进行特征表达，以支持向量机（SVM）作为分类算法，并选取传统的人工神经网络（ANN）作为分类器作为对比，并采用不同大小的训练集和测试集来测试SVM及ANN的分类效果，观察训练集和测试集大小对于分类效果的影响。
|
| 关 键 词： | 垃圾邮件 邮件过滤 支持向量机 人工神经网络 |
| 文章编号： | 1008-9764（2008）01-0100-04 |
| 收稿时间： | 2007-12-05 |
| 修稿时间： | 2007年12月5日 |
Chinese Spam Filtering Based on N-gram and SVM |
| |
| XIA Cheng-feng.Chinese Spam Filtering Based on N-gram and SVM[J].Journal of Guangdong Radio & Television University,2008,17(1):100-103. |
| |
| Authors: | XIA Cheng-feng |
| |
| Institution: | XIA Cheng-feng (Department of Scientific Research,Zhongkai University of Agriculture , Technology,Guangzhou,Guangdong,China,510225) |
| |
| Abstract: | Character expression and text categorization arithmetic are very important in the spam filtering based on content.N-gram is used to express character and Support Vector Machines(SVM) is used as classification algorithm;artificial neural network(ANN) is chosen as the contract,and different sizes of training set and testing set is used to test the results of SVM and ANN classifications.The effects of size training set and testing set are also observed. |
| |
| Keywords: | spare spare filter SVM ANN |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
