CLEF 2003 Experiments at UB: Automatically Generated Phrases and Relevance Feedback for Improving CLIR
This paper presents the results obtained by the University at Buffalo (UB) in CLEF 2003. Our efforts concentrated in the monolingual retrieval and large multilingual retrieval tasks. We used a modified version of the SMART system, a heuristic method based on bigrams to generate phrases that works across multiple languages, and pseudo relevance feedback. Query translation was performed using publicly available machine translation software. Our results show small but consistent improvements in performance due to the use of bigrams. We also found that pseudo relevance feedback benefits from using these bigrams for expanding queries in all the 8 languages that we tested.
Unable to display preview. Download preview PDF.
- 1.Oard, D.: Adaptive Vector Space Text Filtering for Monolingual and Cross-Language Applications. PhD thesis, University of Maryland (1996)Google Scholar
- 2.Porter, M.F.: An algorithm for suffix stripping. Program 14, 130–137 (1980)Google Scholar
- 3.Powel, A.T., French, J.C., Callan, J., Connell, M., Viles, C.L.: The impact of database selection on distributed searching. In: Belkin, N., Ingwersen, P., Leong, M. (eds.) Proceedings of the 23rd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 232–239. ACM Press, New York (2000)CrossRefGoogle Scholar
- 4.Salton, G. (ed.): The SMART Retrieval System: Experiments in Automatic Document Processing. Prentice-Hall, Englewood Cliffs (1983)Google Scholar
- 5.Savoy, J.: Report on CLEF 2002 experiments: Combining multiple sources of evidence. In: Peters, C. (ed.) Results of the CLEF 2002 Cross-Language System Evaluation Campaign: Working Notes for the CLEF 2002 Workshop (2002)Google Scholar