Retrieving attributes using web tables

In this paper we propose an attribute retrieval approach which extracts and ranks attributes from Web tables. We use simple heuristics to filter out improbable attributes and we rank attributes based on frequencies and a table match score. Ranking is reinforced with external evidence from Web search, DBPedia and Wikipedia. Our approach can be applied to whatever instance (e.g. Canada) to retrieve its attributes (capital, GDP). It is shown it has a much higher recall than DBPedia and Wikipedia and that it works better than lexico-syntactic rules for the same purpose. Categories and Subject Descriptors H.3.3 [Information Systems]: Information Search and Retrieval General Terms Experimentation, Performance Keywords information retrieval, attribute retrieval, entity-oriented search

• Type: Journal
• Year: 2011
• Where: JCDL
• Authors: Arlind Kopliku, Karen Pinel-Sauvagnat, Mohand Boughanem
• Keywords: Capital Gdp, Education, JCDL 2011, Oriented Search, Web Tables