"Linked data as background knowledge for information extraction on the web" by Ziqi Zhang, Anna Lisa Gentile and Isabelle Augenstein with Martin Vesely as coordinator

Ziqi Zhang, Anna Lisa Gentile, Isabelle Augenstein

Abstract: Information Extraction (IE) is the technique for transforming textual data into structured representation that can be understood by machines. It is a crucial technique in enabling the Semantic Web, where increasing interest has been seen in recent years. This article reports recent progress in the LODIE project - Linked Open Data for Information Extraction, aimed at advancing Web IE to a new frontier by exploiting largely available, semantically annotated, Linked Open Data as background knowledge. We cover topics of wrapper induction, IE from semi-structured content such as tables and lists, and IE from free-text. We describe new challenges in the research and methods proposed to address them, together with summaries of recent evaluations showing encouraging results.