跳到主要导航 跳到搜索 跳到主要内容

Automatic extraction rules generation based on XPath pattern learning

  • East China Normal University

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

Web forums have become important information sources on the Web due to their rich content contributed by millions of Internet users every day. Data extraction from Web pages is a key but cumbersome step for data analysis because of significant human intervention. Web forums have fairly regular structures which allow us to generate extraction rules automatically according to their paths. In this paper, we introduce formal expressions for XPath patterns and pattern mapping rules, and advise machine learning methods to generate extraction rules for automatic data extraction from Web forums. The experimental results on real-life Web forums show good feasibility and accuracy for forum data.

源语言英语
主期刊名Web Information Systems Engineering - WISE 2010 Workshops - WISE 2010 International Symposium WISS and International Workshops CISE, MBC, Revised Selected Papers
出版商Springer Verlag
58-69
页数12
ISBN(印刷版)9783642243950
DOI
出版状态已出版 - 2011
活动Workshops on Web Information Systems Engineering, WISE 2010: 1st International Symposium on Web Intelligent Systems and Services, WISS 2010, 2nd International Workshop on Mobile Business Collaboration, MBC 2010 and 1st International Workshop on Cloud Information System Engineering, CISE 2010 - Hong Kong, 中国
期限: 12 12月 201014 12月 2010

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
6724 LNCS
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议Workshops on Web Information Systems Engineering, WISE 2010: 1st International Symposium on Web Intelligent Systems and Services, WISS 2010, 2nd International Workshop on Mobile Business Collaboration, MBC 2010 and 1st International Workshop on Cloud Information System Engineering, CISE 2010
国家/地区中国
Hong Kong
时期12/12/1014/12/10

指纹

探究 'Automatic extraction rules generation based on XPath pattern learning' 的科研主题。它们共同构成独一无二的指纹。

引用此