An assessment of open relation extraction systems for the semantic web

Abstract

Open relation extraction has been a growing field of research in the last few years. This paper compares some of the most prominent open relation extractors and explores their strength and weaknesses on standard datasets. In particular, we highlight the lack of formal guidelines that define a valid relation and state that open relation extractors must be situated in particular tasks and contexts. In this respect, we briefly analyze the role of open relation extractors for the semantic Web and linked data context.