University of Arizona at SemEval-2019 Task 12: Deep-Affix Named Entity Recognition of Geolocation Entities

Proceedings of the 13th International Workshop on Semantic Evaluation (SemEval-2019), pages 1319–1323 Minneapolis, Minnesota, USA, June 6–7, 2019. ©2019 Association for Computational Linguistics 1319 University of Arizona at SemEval-2019 Task 12: Deep-Affix Named Entity Recognition of Geolocation Entities Vikas Yadav, Egoitz Laparra, Ti-Tai Wang, Mihai Surdeanu, Steven Bethard University of Arizona {vikasy,laparra,twang03,msurdeanu,bethard}@email.arizona.edu Abstract We present the Named Entity Recognition (NER) and disambiguation model used by the University of Arizona team (UArizona) for SemEval 2019 task 12. We achieved fourth place on tasks 1 and 3. We implemented a deep-affix based LSTM-CRF NER model for task 1, which utilizes only character, word, pre- fix and suffix information for the identification of geolocation entities. Despite using just the training data provided by task organizers and not using any lexicon features, we achieved 78.85% strict micro F-score on task 1. We used the unsupervised population heuristics for task 3 and achieved 52.99% strict micro-F1 score in this task. 1 Introduction Geoparsing is the task of detecting geolocation phrases in unstructured text and normalizing them to a unique identifier, e.g. GeoNames1 IDs. Al- though many automatic resolvers have been re- leased in the past years, their performance fluc- tuates when applied to different domains (Gritta et al., 2018b). Most have also not been applied to and evaluated on scientific publications. The Sem- Eval 2019 Shared Task 12: Toponym Resolution in Scientific Papers (Weissenbacher et al., 2019) aims to boost the research on geoparsing for the scien- tific domain by focusing on epidemiology journal articles. The task includes three sub-tasks: toponym de- tection, toponym disambiguation, and end-to-end toponym resolution. The first one requires par- ticipants to detect the text boundaries of all to- ponym mentions in articles. In toponym disam- biguation, the toponym mentions are known, and the resolver has to align them to their precise coor- dinates through GeoNames IDs. For the last sub- 1http://www.geonames.org/ task, the resolver must perform both detection and disambiguation. In this paper, we present the description of our system for SemEval 2019 Shared Task 12, in which we focus mainly on toponym detection. For this sub-task, we propose a recurrent neural network that combines word, character and affix informa- tion. By making use of the baseline provided by the organizers for toponym disambiguation, we also obtain results for the end-to-end sub-task. 2 Related Work Toponym detection and resolution has been widely studied, and various systems (Gritta et al., 2018b) have been proposed for these tasks. Toponym de- tection has been implemented on texts from various sources like social media (Karagoz et al., 2016), PubMed articles (Magge et al., 2018) etc. Various named entity recognition (NER) systems including rule-based (Gritta et al., 2018b), machine learning- based (Karagoz et al., 2016), and deep learning- based (Magge et al., 2018) have been implemented for detecting toponyms. The disambiguation step has been tackled pre- viously using both supervised models and unsu- pervised heuristic based approaches. For example, Turton (2008) presented a rule based system for disambiguating locations from PubMed abstracts. Weissenbacher et al. …

Content truncated to 500 words