Joint entity recognition and relation extraction as a multi-head selection problem

• Published in: Expert systems with applications Vol. 114; pp. 34 - 45
• Main Authors: Bekoulis, Giannis, Deleu, Johannes, Demeester, Thomas, Develder, Chris
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 30.12.2018; Elsevier BV
• Subjects: Data models; Dependence; Entity recognition; Feature extraction; Head; Joint model; Multi-head selection; Natural language processing; Parsers; Pattern recognition; Random variables; Real estate; Recognition; Relation extraction; Sequence labeling; Relation extraction; Entity recognition; Sequence labeling; Joint model; Multi-head selection
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2018.07.032

•We propose a new joint neural model for entity recognition and relation extraction.•Our model does not rely on external NLP tools nor hand-crafted features.•Entities and relations within the same sentence are extracted simultaneously.•We experimentally evaluate extensively on four datasets (ACE04, CoNLL04, ADE, DREC).•Our model achieves state-of-the-art performance in different contexts and languages. State-of-the-art models for joint entity recognition and relation extraction strongly rely on external natural language processing (NLP) tools such as POS (part-of-speech) taggers and dependency parsers. Thus, the performance of such joint models depends on the quality of the features obtained from these NLP tools. However, these features are not always accurate for various languages and contexts. In this paper, we propose a joint neural model which performs entity recognition and relation extraction simultaneously, without the need of any manually extracted features or the use of any external tool. Specifically, we model the entity recognition task using a CRF (Conditional Random Fields) layer and the relation extraction task as a multi-head selection problem (i.e., potentially identify multiple relations for each entity). We present an extensive experimental setup, to demonstrate the effectiveness of our method using datasets from various contexts (i.e., news, biomedical, real estate) and languages (i.e., English, Dutch). Our model outperforms the previous neural models that use automatically extracted features, while it performs within a reasonable margin of feature-based neural models, or even beats them.