A comparison of methods for the automatic identification of locations in wikipedia

ABSTRACT

In this paper we compare two methods for the automatic identification of geographical articles in encyclopedic resources such as Wikipedia. The methods are a WordNet-based method that uses a set of keywords related to geographical places, and a multinomial Naïve Bayes classificator, trained over a randomly selected subset of the English Wikipedia. This task may be included into the broader task of Named Entity classification, a well-known problem in the field of Natural Language Processing. The experiments were carried out considering both the full text of the articles and only the definition of the entity being described in the article. The obtained results show that the information contained in the page templates and the category labels is more useful than the text of the articles.

REFERENCES

• 1: D. Buscaldi, P. Rosso, and P. Peris. Inferring geographical ontologies from multiple resources for geographical information retrieval. In C. Jones and R. Purves, editors, Proceedings of 3rd SIGIR Workshop on Geographical Information Retrieval, August 2006.
• 2: D. Buscaldi, P. Rosso, and E. Sanchis. Wordnet as a geographical information resource. In Proceedings of the 3rd Global WordNet Association (GWA06), 2006.
• 3: S. Cucerzan. Large scale named entity disambiguation based on wikipedia data. In The EMNLP-CoNLL Joint Conference, 2007.
• 4: Ludovic Denoyer , Patrick Gallinari, The Wikipedia XML corpus, ACM SIGIR Forum, v.40 n.1, June 2006 [doi>10.1145/1147197.1147210]
• 5: G. Fu, C. B. Jones, and A. I. Abdelmoty. Bulding a geographical ontology for intelligent spatial search on the web. In Proceedings of the IASTED International Conference on Databases and Applications, 2005.
• 6: Christopher B. Jones , R. Purves , A. Ruas , M. Sanderson , M. Sester , M. van Kreveld , R. Weibel, Spatial information retrieval and geographical ontologies an overview of the SPIRIT project, Proceedings of the 25th annual international ACM SIGIR conference on Research and development in information retrieval, August 11-15, 2002, Tampere, Finland [doi>10.1145/564376.564457]
• 7: B. Martins, M. Chaves, and M. J. Silva. Assigning geographical scopes to web pages. In Advances in Information Retrieval, volume 3408 of Lecture Notes in Computer Science, pages 564--567. Springer, Berlin, 2005.
• 8: George A. Miller, WordNet: a lexical database for English, Communications of the ACM, v.38 n.11, p.39-41, Nov. 1995 [doi>10.1145/219717.219748]
• 9: S. Overell and S. Rüger. Identifying and grounding descriptions of places. In C. Jones and R. Purves, editors, Proceedings of the 3rd SIGIR Workshop on Geographic Information Retrieval, pages 14--16, August 2006.
• 10: D. Pinto, H. Jiménez-Salazar, P. Rosso, and E. Sanchis. Buap-upv tpirs: A system for document indexing reduction at webclef. In S. Verlag, editor, Accessing Multilingual Information Repositories, Revised Selected Papers CLEF05, volume 4022, pages 873--879, 2006.
• 11: Salim Roukos, Language representation, Survey of the state of the art in human language technology, Cambridge University Press, New York, NY, 1997
• 12: M. Sanderson and J. Kohler. Analyzing geographic queries. In Proceedings of the 1st SIGIR Workshop on Geographic Information Retrieval, 2004.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.7 Natural Language Processing
          Subjects: Text analysis

General Terms:
Experimentation, Measurement

Keywords:
geographical information retrieval, named entity classification