A Joint Named-Entity Recognizer for Heterogeneous Tag-sets Using a Tag Hierarchy

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 140–150 Florence, Italy, July 28 - August 2, 2019. c⃝2019 Association for Computational Linguistics 140 A Joint Named-Entity Recognizer for Heterogeneous Tag-sets Using a Tag Hierarchy Genady Beryozkin, Yoel Drori, Oren Gilon, Tzvika Hartman and Idan Szpektor Google Research Tel Aviv, Israel {genady,dyoel,ogilon,tzvika,szpektor} @google.com Abstract We study a variant of domain adaptation for named-entity recognition where multiple, het- erogeneously tagged training sets are avail- able. Furthermore, the test tag-set is not iden- tical to any individual training tag-set. Yet, the relations between all tags are provided in a tag hierarchy, covering the test tags as a com- bination of training tags. This setting occurs when various datasets are created using differ- ent annotation schemes. This is also the case of extending a tag-set with a new tag by an- notating only the new tag in a new dataset. We propose to use the given tag hierarchy to jointly learn a neural network that shares its tagging layer among all tag-sets. We compare this model to combining independent models and to a model based on the multitasking ap- proach. Our experiments show the benefit of the tag-hierarchy model, especially when fac- ing non-trivial consolidation of tag-sets. 1 Introduction Named Entity Recognition (NER) has seen signif- icant progress in the last couple of years with the application of Neural Networks to the task. Such models achieve state-of-the-art performance with little or no manual feature engineering (Collobert et al., 2011; Huang et al., 2015; Lample et al., 2016; Ma and Hovy, 2016; Dernoncourt et al., 2017). Following this success, more complex NER setups are approached with neural models, among them domain adaptation (Qu et al., 2016; He and Sun, 2017; Dong et al., 2017). In this work we study one type of domain adap- tation for NER, denoted here heterogeneous tag- sets. In this variant, samples from the test set are not available at training time. Furthermore, the test tag-set differs from each training tag-set. However every test tag can be represented either as a single training tag or as a combination of several train- ing tags. This information is given in the form of a hypernym hierarchy over all tags, training and test (see Fig. 1). This setting arises when different schemes are used for annotating multiple datasets for the same task. This often occurs in the medical domain, where healthcare providers use customized tag- sets to create their own private test sets (Shickel et al., 2017; Lee et al., 2018). Another scenario is selective annotation, as in the case of extend- ing an existing tag-set, e.g. {‘Name’, ‘Location’}, with another tag, e.g. ‘Date’. To save annotation effort, new training data is labeled only with the new tag. This case of disjoint tag-sets is also dis- cussed in the work of Greenberg et al. (2018). A similar case is extending a training-set with new examples in which only rare tags are annotated. In domains where training data is scarce, …

Content truncated to 500 words