Identifying Geographical Entities in Users’ Queries
In 2009 we built a system in order to compete in the LAGI task (Log Analysis and Geographic Query Identification). The system uses an external resource built into GATE in combination with Wikipedia and Tumba in order to identify geographical entities in user’s queries. The results obtained with and without Wikipedia resources are comparable. The main advantage of only using GATE resources is the improved run time. In the process of system evaluation we have identified the main problem of our approach: the system has insufficient external resources for the recognition of geographic entities.
KeywordsExternal Resource User Query Correct Match Geographical Entity Resource Loading
Unable to display preview. Download preview PDF.
- 1.Cunningham, H., Maynard, D., Bontcheva, K., Tablan, V.: GATE: an architecture for development of robust HLT applications. In: ACL 2002: Proceedings of the 40th Annual Meeting on Association for Computational Linguistics, Morristown, NJ, USA, pp. 168–175. Association for Computational Linguistics (2001)Google Scholar
- 2.Mandl, T., Agosti, M., Di Nunzio, G.M., Yeh, A., Mani, I., Doran, C., Schulz, J.M.: LogCLEF 2009: the CLEF 2009 Multilingual Logfile Analysis Track Overview. In: Proceedings of the CLEF 2009 Workshop, Corfu, Greece, September 30 - October 2 (2009)Google Scholar