基于结构信息和时域信息的垃圾网页检测技术  

Web Spam detection based on structural and temporal information

作　　者：武磊[1] 高斌[2] 李京[1] 

机构地区：[1]中国科学院软件研究所软件工程技术中心,北京100080 [2]微软亚洲研究院互联网搜索与数据挖掘组,北京100080

出　　处：《计算机应用研究》2008年第4期1243-1246,共4页Application Research of Computers

摘　　要：提出一种将结构信息和时域信息综合运用来检测Spam技术的方法,并针对目前流行的Spam技术,从四个月的网络链接图中提取大量的结构和时域信息特征,训练一组检测Spam技术的网页分类器,取得了很好的实验结果。A novel framework was proposed to combine structural information and temporal information together for Spam detection. Targeting at detecting Some popularity spamming technologies, a group of well-designed structural and temporal features was extracted from a series of link graphs of four successive months, and train a set of classifiers to distinguish normal websites from spam websites. Experiments on a real-world dataset show that the proposed method is quite effective for Web Spam detection

关 键 词：Spam技术 搜索引擎优化 PAGERANK 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]