一种利用相邻页面信息修正分类结果的方法

下载PDF

导出

摘要提出一种利用相邻网页信息修正分类结果的方法。这种方法首先利用NaiveBayes分类器分类待分类网页;其次对待分类网页做链接分析并计算页面与链出页面的相似度,对初始分类结果做修正,得到最终分类结果。根据该方法实现了网页自动分类原型系统,并进行分类实验,实验表明该方法有效提高了分类性能。

作者郭淼霞吴扬扬

机构地区华侨大学计算机系

出处《福建电脑》 2008年第4期78-79,共2页 Journal of Fujian Computer

关键词网页分类相邻网页链接 NAIVE BAYES

参考文献7

1冯是聪,单松巍,龚笔宏,张志刚,李晓明.“天网”目录导航服务研究[J].计算机研究与发展,2004,41(4):653-659. 被引量：8
2Dou Shen;Sun Jiantao;Qiang Yang.A comparison of implicit and explicit links for Web page classification,2006.
3Joachims T;Cristianini N;Shawe-Taylor J.Composite kemels for hypertext categorisation,2001.
4Shih LK;Karger DR.Using URLs and table layout for Web classification tasks,2004.
5Glover EJ;Tsioutsiouliklis K;Lawrence S;Pennock DM,Flake GW.Using web structure for classifying and describing Web pages,2002.
6Yang Y;Alattery S;Ghani R.A study of approaches to hypertext categorization,2002(2-3).
7Oh HJ;Myaeng SH;Lee MH.A practical hypertext categorization method using links and incrementally available class information,2000.

1WebInfomallWebsitshttp://net.cs.pku.edu.cn/-webg/infomall/index.html . 2002
2TianwangsearchengineWebsits http://e.pku.edu.cn . 1997
3http://cn.yahoo.com . 2003
4YYang,XLiu.Are examinationoftextcategorizationmethods[].ACMSIGIRConfonResearchandDevelopmentinInformationRetrieval.1999
5FengShicong,ShanSongwei,ZhangZhigongetal.AdatasetofChineseWebpagesanditscategorization[].ProcoftheCross straitInformationTechnologyWorkshop.2002
6YYang,JanOPedersen.Acomparativestudyonfeatureselectionintextcategorization[].ThethInt’’lConfonMachineLearning.1997
7YYang.Astudyonthresholdingstrategiesfortextcategoriza tion[].ACMSIGIRConfonResearchandDevelopmentinInforma tionRetrieval.2001
8SChakrabarti.Dataminingforhypertext:Atutorialsurvey[].ACMSIGKDDExplorations.2000
9LeiMing,WangJianyong,ChenBaojueetal.Improvedrele vancerankinginwebgather[].JournalofComputerScienceandTechnology.2001
10WangJianyong,ShanSongwei,LeiMingetal.Websearchen gine:Characteristicsofuserbehaviorsandtheirimplication[].Sci enceinChinaSeriesF.2001

福建电脑

2008年第4期

内容加载中请稍等...