摘要: In this work we attempted to determine the relative importance of geographical and WordNet-extracted terms with respect remainder query. Our system is based on Lucene uses LingPipe for Named Entity recognition. Geographical are expanded WordNet holonyms synonyms indexed separately. We checked by boosting them reduction factors (0.75, 0.5 0.25). The comparison clean (using only Lucene) shows that it possible improve mean average precision if equal or less than half content words in also observed may help improving recall but term expansion sensible ambigue place names. As a further work, will need implement toponym disambiguation method order reduce impact kind ambiguity.
cnr.it 参考文章(7)
Davide Buscaldi, Paolo Rosso, Emilio Sanchis Arnal, Using the wordnet ontology in the GeoCLEF geographical information retrieval task cross language evaluation forum. pp. 939- 946 ,(2005) , 10.1007/11878773_105
Davide Buscaldi, Paolo Rosso, Emilio Sanchis, A wordnet-based indexing technique for geographical information retrieval cross language evaluation forum. pp. 954- 957 ,(2006) , 10.1007/978-3-540-74999-8_122
Hideo Joho, Paul D. Clough, Ray R. Larson, Fredric C. Gey, Mark Sanderson, GeoCLEF: the CLEF 2005 Cross-Language Geographic Information Retrieval Track. CLEF (Working Notes). ,(2005)
Eric Garbin, Inderjeet Mani, Disambiguating toponyms in news Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing - HLT '05. pp. 363- 370 ,(2005) , 10.3115/1220575.1220621
George A. Miller, WordNet Communications of the ACM. ,vol. 38, pp. 39- 41 ,(1995) , 10.1145/219717.219748
Davide Buscaldi, Paulo Rosso, A conceptual density-based approach for the disambiguation of toponyms International Journal of Geographical Information Science. ,vol. 22, pp. 301- 313 ,(2008) , 10.1080/13658810701626251
Simon E Overell, Joao Magalhaes, Stefan M Rüger, Place disambiguation with co-occurrence models. CLEF (Working Notes). ,(2006)