Jointly Extracting Relations with Class Ties via Effective Deep Ranking

Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 1810–1820 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1166 Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 1810–1820 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1166 Jointly Extracting Relations with Class Ties via Effective Deep Ranking Hai Ye1, Wenhan Chao1, Zhunchen Luo2∗, Zhoujun Li1 1School of Computer Science and Engineering, Beihang University, Beijing 100191, China {yehai, chaowenhan, lizj}@buaa.edu.cn 2China Defense Science and Technology Information Center, Beijing 100142, China zhunchenluo@gmail.com Abstract Connections between relations in relation extraction, which we call class ties, are common. In distantly supervised scenario, one entity tuple may have multiple relation facts. Exploiting class ties between rela- tions of one entity tuple will be promising for distantly supervised relation extrac- tion. However, previous models are not ef- fective or ignore to model this property. In this work, to effectively leverage class ties, we propose to make joint relation extrac- tion with a unified model that integrates convolutional neural network (CNN) with a general pairwise ranking framework, in which three novel ranking loss functions are introduced. Additionally, an effective method is presented to relieve the severe class imbalance problem from NR (not re- lation) for model training. Experiments on a widely used dataset show that leverag- ing class ties will enhance extraction and demonstrate the effectiveness of our model to learn class ties. Our model outperforms the baselines significantly, achieving state- of-the-art performance. 1 Introduction Relation extraction (RE) aims to classify the relations between two given named entities from natural-language text. Supervised machine learn- ing methods require numerous labeled data to work well. With the rapid growth of volume of relation types, traditional methods can not keep up with the step for the limitation of labeled data. In order to narrow down the gap of data spar- sity, Mintz et al. (2009) propose distant supervi- sion (DS) for relation extraction, which automati- ∗ Corresponding author. place lived (Patsy Ramsey, Atlanta) place of birth (Patsy Ramsey, Atlanta) Sentence Latent Label #1 Patsy Ramsey has been living in Atlanta since she was born. place of birth #2 Patsy Ramsy always loves At- lanta since it is her hometown. place lived Table 1: Training instances generated by freebase. cally generates training data by aligning a knowl- edge facts database (ie. Freebase (Bollacker et al., 2008)) with texts. Class ties mean the connections between rela- tions in relation extraction. In general, we con- clude that class ties can have two types: weak class ties and strong class ties. Weak class ties mainly involve the co-occurrence of relations such as place of birth and place lived, CEO of and founder of. On the contrary, strong class ties mean that relations have latent logical entailments. Take the two relations of capital of and city of for example, if one entity tuple has the rela- tion of capital of, it must express the relation fact …

Content truncated to 500 words