Bm25 Variant Models
BM25 Variant Models is a probabilistic information retrieval framework that extends the original BM25 ranking function by modifying its parameters, normalization strategies, or term weighting schemes to improve document ranking performance under specific retrieval conditions.
16
Publications
4K
Citations
47
Authors
23
Institutions
Publications per year
2017–2026
5
Parents
16
When documents are very long, BM25 fails!
Yuanhua Lv, ChengXiang Zhai · 2011 · 92 citations
Using BM25F for semantic search
José R. Pérez-Agüera, Javier Arroyo, Jane Greenberg et al. · 2010 · 77 citations
Integrating the Probabilistic Models BM25/BM25F into Lucene
Joaquín Pérez-Iglesias, José R. Pérez-Agüera, Víctor Fresno et al. · ArXiv.org · 2009 · 71 citations · Full text
1–5 of 16
47
3
h-index
3
h-index
2
h-index
2
h-index
2
h-index
1–5 of 47
23
Microsoft Research (United Kingdom)
Cambridge, United Kingdom
2 h-index
4 publications
5K citations

York University
Toronto, Canada
2 h-index
6 publications
99 citations

Centre National de la Recherche Scientifique
Paris, France
2 h-index
3 publications
71 citations
Université Mohammed VI Polytechnique
Marrakesh, Morocco
1 h-index
1 publications
17 citations

Université Jean Monnet
Saint-Etienne, France
1 h-index
2 publications
54 citations
Université Claude Bernard Lyon 1
Villeurbanne, France
1 h-index
2 publications
54 citations
Université Ibn-Tofail
Kenitra, Morocco
1 h-index
2 publications
34 citations

National Institute of Informatics
Tokyo, Japan
1 h-index
2 publications
50 citations
1–8 of 23
Conference · 1 publications · 44 citations
1
h-index
1 publications · 17 citations
1
h-index
1 publications · 13 citations
1
h-index
1 publications · 2.9K citations
1
h-index
Repository · 1 publications · 71 citations
1
h-index
1–5