Entity relation discovery from web tables and links

Cindy Xide Lin, Bo Zhao, Tim Weninger, Jiawei Han, Bing Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The World-Wide Web consists not only of a huge number of unstructured texts, but also a vast amount of valuable structured data. Web tables [2] are a typical type of structured information that are pervasive on the web, and Web-scale methods that automatically extract web tables have been studied extensively [1]. Many powerful systems (e.g.OCTOPUS [4], Mesa [3]) use extracted web tables as a fundamental component. In the database vernacular, a table is defined as a set of tuples which have the same attributes. Similarly, a web table is defined as a set of rows (corresponding to database tuples) which have the same column headers (corresponding to database attributes). Therefore, to extract a web table is to extract a relation on the web. In databases, tables often contain foreign keys which refer to other tables. Therefore, it follows that hyperlinks inside a web table sometimes function as foreign keys to other relations whose tuples are contained in the hyperlink's target pages. In this paper, we explore this idea by asking: can we discover new attributes for web tables by exploring hyperlinks inside web tables? This poster proposes a solution that takes a web table as input. Frequent patterns are generated as new candidate relations by following hyperlinks in the web table. The confidence of candidates are evaluated, and trustworthy candidates are selected to become new attributes for the table. Finally, we show the usefulness of our method by performing experiments on a variety of web domains.

Original languageEnglish (US)
Title of host publicationProceedings of the 19th International Conference on World Wide Web, WWW '10
Pages1145-1146
Number of pages2
DOIs
StatePublished - Jul 20 2010
Event19th International World Wide Web Conference, WWW2010 - Raleigh, NC, United States
Duration: Apr 26 2010Apr 30 2010

Publication series

NameProceedings of the 19th International Conference on World Wide Web, WWW '10

Other

Other19th International World Wide Web Conference, WWW2010
Country/TerritoryUnited States
CityRaleigh, NC
Period4/26/104/30/10

Keywords

  • entity relation discovery
  • link
  • web table

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Entity relation discovery from web tables and links'. Together they form a unique fingerprint.

Cite this