Fusion of Overlapped Result Sets
The existence of replicated documents or indexes in dis- tributed information retrieval introduces some level of overlapping be- tween its component databases. In this paper, a new data fusion method taking in account the overlapping information is compared with a conven- tional fusion method. The overall system effectiveness evaluation enables preliminary conclusions about the importance of such parameter for data fusion in distributed information retrieval.
KeywordsDistributed Information Retrieval Replication Data Fusion
Unable to display preview. Download preview PDF.
- 1.J. Ho Lee. Analyses of Multiple Evidence Combination. In Proc. of ACM SIGIR Conference on Research in Information Retrieval, Philadelphia, USA, 1997.Google Scholar
- 2.Z. Lu, J. Callan, and W. B. Croft. Applying Inference Networks to Multiple Collections Searching. Technical report, CS, Univ. of Massachusets, 1996.Google Scholar
- 3.J. Macedo, A. Costa, and V. Freitas. Webtrieve: A distributed retrieval architecture for Text Databases.Technical report, Universidade do Minho, 1998.URL em ftp://marco.uminho.pt/Docs/local/RT/webtrieve. ps (in Portuguese).
- 4.L. Gravano, K. Chang, H. Garcia-Molina, C. Lagoze, and A. Paepcke. STARTS, Stanford Protocol Proposal for Internet Retrieval and Search. Technical report, DL Project, Stanford University, Jan 1997.Google Scholar
- 5.N.J. Belkin and P. Kantor. Combining the Evidence of Multiple Query Representations for Information Retrieval.IP & M, 31(3):431–448, 1995.Google Scholar