Going out on a limb: Joint Extraction of Entity Mentions and Relations without Dependency Trees

Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 917–928 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1085 Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 917–928 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1085 Going out on a limb: Joint Extraction of Entity Mentions and Relations without Dependency Trees Arzoo Katiyar and Claire Cardie Department of Computer Science Cornell University Ithaca, NY, 14853, USA arzoo, cardie@cs.cornell.edu Abstract We present a novel attention-based re- current neural network for joint extrac- tion of entity mentions and relations. We show that attention along with long short term memory (LSTM) network can extract semantic relations between entity men- tions without having access to dependency trees. Experiments on Automatic Content Extraction (ACE) corpora show that our model significantly outperforms feature- based joint model by Li and Ji (2014). We also compare our model with an end-to- end tree-based LSTM model (SPTree) by Miwa and Bansal (2016) and show that our model performs within 1% on entity mentions and 2% on relations. Our fine- grained analysis also shows that our model performs significantly better on AGENT- ARTIFACT relations, while SPTree per- forms better on PHYSICAL and PART- WHOLE relations. 1 Introduction Extraction of entities and their relations from text belongs to a very well-studied family of structured prediction tasks in NLP. There are several NLP tasks such as fine-grained opinion mining (Choi et al., 2006), semantic role labeling (Gildea and Jurafsky, 2002), etc., which have a similar struc- ture; thus making it an important and a challeng- ing task. Several methods have been proposed for entity mention and relation extraction at the sentence- level. These can be broadly categorized into – 1) pipeline models that treat the identification of entity mentions (Nadeau and Sekine, 2007) and relation classification (Zhou et al., 2005) as two separate tasks; and 2) joint models, also the more recent, which simultaneously identify the entity mention and relations (Li and Ji, 2014; Miwa and Sasaki, 2014). Joint models have been argued to perform better than the pipeline models as knowl- edge of the typed relation can increase the confi- dence of the model on entity extraction and vice versa. Recurrent networks (RNNs) (Elman, 1990) have recently become very popular for sequence tagging tasks such as entity extraction that in- volves a set of contiguous tokens. However, their ability to identify relations between non-adjacent tokens in a sequence, e.g., the head nouns of two entities, is less explored. For these tasks, RNNs that make use of tree structures have been deemed more suitable. Miwa and Bansal (2016), for ex- ample, propose an RNN comprised of a sequence- based long short term memory (LSTM) for en- tity identification and a separate tree-based depen- dency LSTM layer for relation classification using shared parameters between the two components. As a result, their model depends critically on ac- cess to dependency trees, restricting it to …

Content truncated to 500 words