论文信息 - Logistic Regression Merging of Amberfish and Lucene Multisearch Results

Logistic Regression Merging of Amberfish and Lucene Multisearch Results

A simple logistic-regression based isolated data fusion algorithm was used to merge results from two free open-source text retrieval tools. The algorithm is described and results from each search tool are compared against the merged results and against each other. Basic performance measures are reported and discussed, and future projects are outlined.

Gregory B. Newby | Christopher T. Fallen | G. Newby | C. Fallen

[1] Fredric C. Gey,et al. Inferring probability of relevance using the method of logistic regression , 1994, SIGIR '94.

[2] Ellen M. Voorhees,et al. Siemens TREC-4 Report: Further Experiments with Database Merging , 1995, TREC.

[3] Nassib Nassar,et al. Amberfish at the TREC 2004 Terabyte Track , 2004, Text Retrieval Conference.

[4] Otis Gospodnetic,et al. Lucene in Action , 2004 .

[5] Jacques Savoy,et al. Database merging strategy based on logistic regression , 2000, Inf. Process. Manag..

[6] Jacques Savoy,et al. Report on the TREC-5 Experiment: Data Fusion and Collection Fusion , 1996, TREC.

[7] David Hawking,et al. Merging Results From Isolated Search Engines , 1999, Australasian Database Conference.