Modeling Joint Entity and Relation Extraction with Table Representation

Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1858–1869, October 25-29, 2014, Doha, Qatar. c⃝2014 Association for Computational Linguistics Modeling Joint Entity and Relation Extraction with Table Representation Makoto Miwa and Yutaka Sasaki Toyota Technological Institute 2-12-1 Hisakata, Tempaku-ku, Nagoya, 468-8511, Japan {makoto-miwa, yutaka.sasaki}@toyota-ti.ac.jp Abstract This paper proposes a history-based struc- tured learning approach that jointly ex- tracts entities and relations in a sentence. We introduce a novel simple and flexible table representation of entities and rela- tions. We investigate several feature set- tings, search orders, and learning meth- ods with inexact search on the table. The experimental results demonstrate that a joint learning approach significantly out- performs a pipeline approach by incorpo- rating global features and by selecting ap- propriate learning methods and search or- ders. 1 Introduction Extraction of entities and relations from texts has been traditionally treated as a pipeline of two sep- arate subtasks: entity recognition and relation ex- traction. This separation makes the task easy to deal with, but it ignores underlying dependencies between and within subtasks. First, since entity recognition is not affected by relation extraction, errors in entity recognition are propagated to re- lation extraction. Second, relation extraction is often treated as a multi-class classification prob- lem on pairs of entities, so dependencies between pairs are ignored. Examples of these dependen- cies are illustrated in Figure 1. For dependencies between subtasks, a Live in relation requires PER and LOC entities, and vice versa. For in-subtask dependencies, the Live in relation between “Mrs. Tsutayama” and “Japan” can be inferred from the two other relations. Figure 1 also shows that the task has a flexible graph structure. This structure usually does not cover all the words in a sentence differently from other natural language processing (NLP) tasks such as part-of-speech (POS) tagging and depen- Mrs. Tsuruyama is from Kumamoto Prefecture in Japan . PER LOC LOC Live_in Located_in Live_in Figure 1: An entity and relation example (Roth and Yih, 2004). Person (PER) and location (LOC) entities are connected by Live in and Located in relations. dency parsing, so local constraints are considered to be more important in the task. Joint learning approaches (Yang and Cardie, 2013; Singh et al., 2013) incorporate these de- pendencies and local constraints in their models; however most approaches are time-consuming and employ complex structures consisting of multi- ple models. Li and Ji (2014) recently proposed a history-based structured learning approach that is simpler and more computationally efficient than other approaches. While this approach is promis- ing, it still has a complexity in search and restricts the search order partly due to its semi-Markov rep- resentation, and thus the potential of the history- based learning is not fully investigated. In this paper, we introduce an entity and relation table to address the difficulty in representing the task. We propose a joint extraction of entities and relations using a history-based structured learning on the table. This table representation simplifies the task into a table-filling problem, and …

Content truncated to 500 words