Distant supervision has become the leading method for training large-scale relation extractors, with nearly universal adoption in recent TAC knowledge-base population competitions. However, there are still many questions about the best way to learn such extractors. In this paper we investigate four orthogonal improvements: integrating named entity linking (NEL) and coreference resolution into argument identification for training and extraction, enforcing type constraints of linked arguments, and partitioning the model by relation type signature. We evaluate sentential extraction performance on two datasets: the popular set of NY Times articles partially annotated by Hoffmann et al. (2011) and a new dataset, called GoReCo, that is comprehensively annotated for 48 common relations. We find that using NEL for argument identification boosts performance over the traditional approach (named entity recognition with string match), and there is further improvement from using argument types. Our best system boosts precision by 44% and recall by 70%.
展开▼
