Disambiguating Toponyms in News

Proceedings of Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing (HLT/EMNLP), pages 363–370, Vancouver, October 2005. c⃝2005 Association for Computational Linguistics Disambiguating Toponyms in News Eric Garbin Inderjeet Mani Department of Linguistics Department of Linguistics Georgetown University Georgetown University Washington, DC 20057, USA Washington, DC 20057, USA egarbin@cox.net im5@georgetown.edu Abstract This research is aimed at the problem of disambiguating toponyms (place names) in terms of a classification derived by merging information from two publicly available gazetteers. To establish the dif- ficulty of the problem, we measured the degree of ambiguity, with respect to a gazetteer, for toponyms in news. We found that 67.82% of the toponyms found in a corpus that were ambiguous in a gaz- etteer lacked a local discriminator in the text. Given the scarcity of human- annotated data, our method used unsuper- vised machine learning to develop disam- biguation rules. Toponyms were automatically tagged with information about them found in a gazetteer. A toponym that was ambiguous in the gazet- teer was automatically disambiguated based on preference heuristics. This automatically tagged data was used to train a machine learner, which disambigu- ated toponyms in a human-annotated news corpus at 78.5% accuracy. 1 Introduction Place names, or toponyms, are ubiquitous in natu- ral language texts. In many applications, including Geographic Information Systems (GIS), it is nec- essary to interpret a given toponym mention as a particular entity in a geographical database or gaz- etteer. Thus the mention “Washington” in “He vis- ited Washington last year” will need to be interpreted as a reference to either the city Wash- ington, DC or the U.S. state of Washington, and “Berlin” in “Berlin is cold in the winter” could mean Berlin, New Hampshire or Berlin, Germany, among other possibilities. While there has been a considerable body of work distinguishing between a toponym and other kinds of names (e.g., person names), there has been relatively little work on resolving which place and what kind of place given a classification of kinds of places in a gazetteer. Disambiguated toponyms can be used in a GIS to highlight a position on a map corresponding to the coordinates of the place, or to draw a polygon rep- resenting the boundary. In this paper, we describe a corpus-based method for disambiguating toponyms. To establish the dif- ficulty of the problem, we began by quantifying the degree of ambiguity of toponyms in a corpus with respect to a U.S. gazetteer. We then carried out a corpus-based investigation of features that could help disambiguate toponyms. Given the scarcity of human-annotated data, our method used unsupervised machine learning to develop disam- biguation rules. Toponyms were automatically tagged with information about them found in a gazetteer. A toponym that was ambiguous in the gazetteer was automatically disambiguated based on preference heuristics. This automatically tagged data was used to train the machine learner. We compared this method with a supervised machine learning approach trained on a corpus annotated and disambiguated by hand. Our investigation targeted toponyms …

Content truncated to 500 words