{"title":"Automatic Maintenance ofWeb Directories using Click-Through Data","authors":"A. Cid, Carlos A. Hurtado, Marcelo Mendoza","doi":"10.1109/ICDEW.2006.36","DOIUrl":null,"url":null,"abstract":"In this paper, we propose a method for the automatic maintenance of a directory of Web documents, based on the analysis of the query logs of a search engine. Our method allows automatic maintenance and continuous improvement of Web directories without the need of human editors. The method finds relevant documents to each topic of the Web directory and estimates the relevance of the documents to the topic from user clicks registered in the query log. We also enrich the Web directory by recommending queries to the topics of the directory. We show experiments to illustrate the effectiveness of our approach.","PeriodicalId":331953,"journal":{"name":"22nd International Conference on Data Engineering Workshops (ICDEW'06)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2006-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"22nd International Conference on Data Engineering Workshops (ICDEW'06)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICDEW.2006.36","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 7
Abstract
In this paper, we propose a method for the automatic maintenance of a directory of Web documents, based on the analysis of the query logs of a search engine. Our method allows automatic maintenance and continuous improvement of Web directories without the need of human editors. The method finds relevant documents to each topic of the Web directory and estimates the relevance of the documents to the topic from user clicks registered in the query log. We also enrich the Web directory by recommending queries to the topics of the directory. We show experiments to illustrate the effectiveness of our approach.