RGCL-WLV at SemEval-2019 Task 12: Toponym Detection

Proceedings of the 13th International Workshop on Semantic Evaluation (SemEval-2019), pages 1297–1301 Minneapolis, Minnesota, USA, June 6–7, 2019. ©2019 Association for Computational Linguistics 1297 RGCL-WLV at SemEval-2019 Task 12: Toponym Detection Alistair Plum1 ∗, Tharindu Ranasinghe1 ∗, Pablo Calleja2, Constantin Or˘asan1, Ruslan Mitkov1 1Research Group in Computational Linguistics, University of Wolverhampton, UK 2Ontology Engineering Group, Universidad Polit´ecnica de Madrid, ES {a.j.plum, t.d.ranasinghehettiarachchige}@wlv.ac.uk {c.orasan, r.mitkov}@wlv.ac.uk pcalleja@fi.upm.es Abstract This article describes the system submitted by the RGCL-WLV team to the SemEval 2019 Task 12: Toponym resolution in scientific pa- pers. The system detects toponyms using a bootstrapped machine learning (ML) ap- proach which classifies names identified using gazetteers extracted from the GeoNames ge- ographical database. The paper evaluates the performance of several ML classifiers, as well as how the gazetteers influence the accuracy of the system. Several runs were submitted. The highest precision achieved for one of the sub- missions was 89%, albeit it at a relatively low recall of 49%. 1 Introduction Resolving a toponym, a proper name that refers to a real existing location, is a non-trivial task closely related to named entity recognition (NER) (Pisko- rski and Yangarber, 2013). For this reason, us- ing an NER system to detect and assign location tags could seem a good way forward. However, NER systems may not be able to detect whether a name refers to a actual location or not (e.g., Lon- don in London Bus Company). In addition, loca- tion names are usually ambiguous, which means it is crucial that these are disambiguated in order to assign the correct coordinates. While in the past the focus in toponym res- olution has been on rule and gazetteer driven methods (Speriosu and Baldridge, 2013), more recent approaches also consider ML-based tech- niques. DeLozier et al. (2015) describe their ML-based approach, which does not require a gazetteer. The approach calculates the geograph- ical profile of each word, which is refined using Wikipedia statistics, and then fed into an ML clas- sifier. Speriosu and Baldridge (2013) also make ∗The first two authors contributed equally to the paper. use of an ML classifier which is text-driven. Geo- tags of documents are used to automatically gen- erate a training set. Although the two previous ap- proaches used two standard corpora for toponym resolution, consisting of news articles and 19th century civil war texts, there are wide areas of application for toponym resolution. For instance, Ireson and Ciravegna (2010) explore the use in so- cial media, while Lieberman and Samet (2012) at- tempt to analyse news streams. Spitz et al. (2016) have also used an encyclopaedic dataset, compiled from Wikipedia, WordNet and GeoNames. The focus of the SemEval 2019 Task 12 was to- ponym resolution in journal articles from the bio- medical domain (Weissenbacher et al., 2019). The articles that had to be processed were case stud- ies on the epidemiology of viruses, meaning that the developed systems can potentially be used to track viruses. The task was composed of three sub-tasks: (1) toponym detection, followed by …

Content truncated to 500 words