{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T05:30:14Z","timestamp":1790055014139,"version":"4.0.1"},"reference-count":63,"publisher":"Emerald","issue":"1-2","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2009,9,17]]},"abstract":"<jats:p>The Probabilistic Relevance Framework (PRF) is a formal framework for document retrieval, grounded in work done in the 1970\u20131980s, which led to the development of one of the most successful text-retrieval algo\u00acrithms, BM25. In recent years, research in the PRF has yielded new retrieval models capable of taking into account document meta-data (especially structure and link-graph information). Again, this has led to one of the most successful Web-search and corporate-search algo\u00acrithms, BM25F. This work presents the PRF from a conceptual point of view, describing the probabilistic modelling assumptions behind the framework and the different ranking algorithms that result from its application: the binary independence model, relevance feedback mod\u00acels, BM25 and BM25F. It also discusses the relation between the PRF and other statistical models for IR, and covers some related topics, such as the use of non-textual features, and parameter optimisation for models with free parameters.<\/jats:p>","DOI":"10.1561\/1500000019","type":"journal-article","created":{"date-parts":[[2010,1,12]],"date-time":"2010-01-12T10:08:33Z","timestamp":1263290913000},"page":"1-174","source":"Crossref","is-referenced-by-count":2562,"title":["The Probabilistic Relevance Framework: BM25 and Beyond"],"prefix":"10.1108","volume":"4","author":[{"given":"Stephen","family":"Robertson","sequence":"first","affiliation":[{"name":"Microsoft Research , 7 J J Thomson Avenue, Cambridge CB3 0FB,","place":["UK"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hugo","family":"Zaragoza","sequence":"additional","affiliation":[{"name":"Yahoo! Research, Av. Diagonal 177, Barcelona 08028, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"140","published-online":{"date-parts":[[2009,9,17]]},"reference":[{"key":"2026040314280606700_ref001","article-title":"Proceedings of the NIPS 2005 Workshop on Learning to Rank","author":"Agarwal","year":"2005"},{"issue":"4","key":"2026040314280606700_ref002","doi-asserted-by":"crossref","first-page":"357","DOI":"10.1145\/582415.582416","article-title":"Probabilistic models of information retrieval based on measuring the divergence from randomness","volume":"20","author":"Amati","year":"2002","journal-title":"ACM Transactions on Information Systems"},{"key":"2026040314280606700_ref003","first-page":"143","article-title":"Okapi at TREC-5","volume-title":"The Fifth Text Retrieval Conference (TREC-5). NIST Special Publication","author":"Beaulieu","year":"1997"},{"key":"2026040314280606700_ref004","unstructured":"Berghen, F. V.\n          \n          n.d., Trust Region Algorithms. Webpage. http:\/\/www.lemurproject.org"},{"key":"2026040314280606700_ref005","article-title":"CONDOR: A constrained, non-linear, derivative-free parallel optimizer for continuous, high computing load, noisy objective functions","volume-title":"PhD thesis","author":"Berghen","year":"2004"},{"key":"2026040314280606700_ref006","author":"Bishop","year":"2006"},{"key":"2026040314280606700_ref007","first-page":"993","article-title":"Latent dirichlet allocation","volume":"3","author":"Blei","year":"2003","journal-title":"Journal of Machine Learning Research"},{"key":"2026040314280606700_ref008","doi-asserted-by":"crossref","first-page":"471","DOI":"10.1002\/asi.10398","article-title":"A new unified probabilistic model","volume":"55","author":"Bodoff","year":"2004","journal-title":"Journal of the American Society for Information Science and Technology"},{"key":"2026040314280606700_ref009","doi-asserted-by":"crossref","DOI":"10.6028\/NIST.SP.500-266.terabyte-umilano.vigna","article-title":"MG4J at TREC 2005","volume-title":"The Fourteenth Text Retrieval Conference (TREC 2005) Proceedings, NIST Special Publication","author":"Boldi","year":"2005"},{"key":"2026040314280606700_ref010","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1145\/1102351.1102363","article-title":"Learning to rank using gradient descent","volume-title":"Proceedings of the International Conference on Machine Learning (ICML)","author":"Burges","year":"2005"},{"key":"2026040314280606700_ref011","doi-asserted-by":"crossref","first-page":"110","DOI":"10.1145\/195705.195735","article-title":"Some inconsistencies and misidentified modelling assumptions in probabilistic information retrieval","volume":"13","author":"Cooper","year":"1995","journal-title":"ACM Transactions on Information Systems"},{"key":"2026040314280606700_ref012","doi-asserted-by":"crossref","first-page":"472","DOI":"10.1145\/1076034.1076106","article-title":"Relevance weighting for query independent evidence","volume-title":"Proceedings of the 28th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Craswell","year":"2005"},{"key":"2026040314280606700_ref013","doi-asserted-by":"crossref","DOI":"10.6028\/NIST.SP.500-266.enterprise-microsoft.robertson","article-title":"Microsoft Cambridge at TREC-14: Enterprise track","volume-title":"The Fourteenth Text Retrieval Conference (TREC 2005)","author":"Craswell","year":"2005"},{"issue":"4","key":"2026040314280606700_ref014","doi-asserted-by":"crossref","DOI":"10.1145\/299917.299920","article-title":"\u201cIs this document relevant? \u2026 probably\u201d: A survey of probabilistic models in information retrieval","volume":"30","author":"Crestani","year":"1998","journal-title":"ACM Computing Surveys"},{"key":"2026040314280606700_ref015","doi-asserted-by":"crossref","first-page":"285","DOI":"10.1108\/eb026683","article-title":"Using probabilistic models of document retrieval without relevance information","volume":"35","author":"Croft","year":"1979","journal-title":"Journal of Documentation"},{"key":"2026040314280606700_ref016","author":"Feller","year":"1968"},{"issue":"3","key":"2026040314280606700_ref017","doi-asserted-by":"crossref","DOI":"10.1093\/comjnl\/35.3.243","article-title":"Probabilistic Models in Information Retrieval","volume":"35","author":"Fuhr","year":"1992","journal-title":"The Computer Journal"},{"key":"2026040314280606700_ref018","doi-asserted-by":"crossref","first-page":"465","DOI":"10.1145\/62437.62487","article-title":"Information retrieval using a singular value decomposition model of latent semantic structure","volume-title":"Proceedings of the 11th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Furnas","year":"1988"},{"key":"2026040314280606700_ref019","doi-asserted-by":"crossref","first-page":"197","DOI":"10.1002\/asi.4630260402","article-title":"A probabilistic approach to automatic keyword indexing (parts 1 and 2)","volume":"26","author":"Harter","year":"1975","journal-title":"Journal of the American Society for Information Science"},{"key":"2026040314280606700_ref020","doi-asserted-by":"crossref","first-page":"178","DOI":"10.1145\/1008992.1009025","article-title":"Parsimonious language models for information retrieval","volume-title":"Proceedings of the 27th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Hiemstra","year":"2004"},{"key":"2026040314280606700_ref021","doi-asserted-by":"crossref","first-page":"50","DOI":"10.1145\/312624.312649","article-title":"Probabilistic latent semantic indexing","volume-title":"Proceedings of the 22nd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Hofmann","year":"1999"},{"key":"2026040314280606700_ref022","unstructured":"NA\n              NA\n            \n          \n          Indri Homepage\n          Lemur Project\n          NA\n          http:\/\/www.lemurproject.org\/indri"},{"issue":"2","key":"2026040314280606700_ref023","doi-asserted-by":"crossref","first-page":"58","DOI":"10.1145\/1328964.1328974","article-title":"Learning to rank for information retrieval (LR4IR 2007)","volume":"41","author":"Joachims","year":"2007","journal-title":"SIGIR Forum"},{"key":"2026040314280606700_ref024","doi-asserted-by":"crossref","DOI":"10.1145\/383952.383970","article-title":"Document language models, query models, and risk minimization for information retrieval","volume-title":"Proceedings of the 24th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Lafferty","year":"2001"},{"key":"2026040314280606700_ref025","first-page":"1","volume-title":"Language Modelling for Information Retrieval","author":"Lafferty","year":"2003"},{"key":"2026040314280606700_ref026","doi-asserted-by":"crossref","first-page":"120","DOI":"10.1145\/383952.383972","article-title":"Relevance based language models","volume-title":"Proceedings of the 24th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Lavrenko","year":"2001"},{"key":"2026040314280606700_ref027","unstructured":"Lemur Project\n          \n          Lemur Toolkit Homepage\n          Lemur Project\n          NA\n          http:\/\/www.lemurproject.org"},{"issue":"2","key":"2026040314280606700_ref028","doi-asserted-by":"crossref","first-page":"76","DOI":"10.1145\/1480506.1480519","article-title":"Learning to rank for information retrieval (LR4IR 2008)","volume":"42","author":"Li","year":"2008","journal-title":"SIGIR Forum"},{"key":"2026040314280606700_ref029","unstructured":"Apache Software Foundation\n          \n          Lucene Homepage\n          Apache Software Foundation\n          NA\n          http:\/\/lucene.apache.org"},{"issue":"3","key":"2026040314280606700_ref030","doi-asserted-by":"crossref","first-page":"216","DOI":"10.1145\/321033.321035","article-title":"On relevance, probabilistic indexing and information retrieval","volume":"7","author":"Maron","year":"1960","journal-title":"Journal of the ACM"},{"key":"2026040314280606700_ref031","doi-asserted-by":"crossref","first-page":"253","DOI":"10.1145\/1321440.1321478","article-title":"Automatic feature selection in the Markov random field model for information retrieval","volume-title":"Proceedings of the Sixteenth ACM Conference on Information and Knowledge Management","author":"Metzler","year":"2007"},{"key":"2026040314280606700_ref032","doi-asserted-by":"crossref","first-page":"472","DOI":"10.1145\/1076034.1076115","article-title":"A Markov random field model for term dependencies","volume-title":"Proceedings of the 28th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Metzler","year":"2005"},{"key":"2026040314280606700_ref033","volume-title":"Information Retrieval in Practice","author":"Metzler","year":"2009"},{"key":"2026040314280606700_ref034","unstructured":"Universit\u00e0 degli Studi di Milano \u2013 DSI\n          \n          MG4J Homepage\n          Universit\u00e0 degli Studi di Milano \u2013 DSI\n          NA\n          http:\/\/mg4j.dsi.unimi.it"},{"key":"2026040314280606700_ref035","unstructured":"MacFarlane\n              A.\n            \n          \n          Okapi-Pack Homepage\n          City University London\n          NA\n          http:\/\/www.soi.city.ac.uk\/andym\/OKAPI-PACK"},{"key":"2026040314280606700_ref036","article-title":"UCM-Y!R at CLEF 2008 Robust and WSD tasks","volume-title":"CLEF 2008 Workshop","author":"P\u00e9rez-Agu\u0308era","year":"2008"},{"key":"2026040314280606700_ref037","article-title":"Exploiting morphological query structure using genetic optimization","volume-title":"NLDB 2008, LNCS, Springer Verlag","author":"P\u00e9rez-Agu\u0308era","year":"2008"},{"key":"2026040314280606700_ref038","unstructured":"P\u00e9rez-Iglesias, J.\n          \n          BM25 and BM25F Implementation for Lucene\n          Webpage\n          http:\/\/nlp.uned.es\/~jperezi\/Lucene-BM25\n          NA"},{"key":"2026040314280606700_ref039","unstructured":"University of Twente\n          \n          PF-Tijah Homepage\n          University of Twente\n          NA\n          http:\/\/dbappl.cs.utwente.nl\/pftijah"},{"key":"2026040314280606700_ref040","doi-asserted-by":"crossref","first-page":"294","DOI":"10.1108\/eb026647","article-title":"The probability ranking principle in information retrieval","volume":"33","author":"Robertson","year":"1977","journal-title":"Journal of Documentation"},{"key":"2026040314280606700_ref041","doi-asserted-by":"crossref","first-page":"359","DOI":"10.1108\/eb026866","article-title":"On term selection for query expansion","volume":"46","author":"Robertson","year":"1990","journal-title":"Journal of Documentation"},{"key":"2026040314280606700_ref042","doi-asserted-by":"crossref","first-page":"239","DOI":"10.1023\/A:1015702129514","article-title":"Threshold setting and performance optimization in adaptive filtering","volume":"5","author":"Robertson","year":"2002","journal-title":"Information Retrieval"},{"key":"2026040314280606700_ref043","first-page":"108","article-title":"The unified probabilistic model for IR","volume-title":"Proceedings of Research and Development in Information Retrieval","author":"Robertson","year":"1983"},{"key":"2026040314280606700_ref044","article-title":"Relevance weighting of search terms","author":"Robertson","year":"1977","journal-title":"Journal of the American Society for Information Science"},{"key":"2026040314280606700_ref045","first-page":"35","article-title":"Probabilistic models of indexing and searching","volume-title":"Information Retrieval Research (Proceedings of Research and Development in Information Retrieval, Cambridge, 1980)","author":"Robertson","year":"1981"},{"key":"2026040314280606700_ref046","first-page":"232","article-title":"Some Simple Effective Approximations to the 2-Poisson Model for Probabilistic Weighted Retrieval","volume-title":"Proceedings of the 17th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Robertson","year":"1994"},{"key":"2026040314280606700_ref047","first-page":"16","article-title":"On relevance weights with little relevance information","volume-title":"Proceedings of the 20th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Robertson","year":"2007"},{"key":"2026040314280606700_ref048","first-page":"21","article-title":"Okapi at TREC","volume-title":"The First Text Retrieval Conference (TREC-1), NIST Special Publication 500-207","author":"Robertson","year":"1992"},{"issue":"3","key":"2026040314280606700_ref049","doi-asserted-by":"crossref","first-page":"321","DOI":"10.1007\/s10791-007-9025-9","article-title":"On rank-based effectiveness measures and optimization","volume":"10","author":"Robertson","year":"2007","journal-title":"Information Retrieval"},{"key":"2026040314280606700_ref050","first-page":"42","article-title":"Simple BM25 extension to multiple weighted fields","volume-title":"Proceedings of the 2004 ACM CIKM International Conference on Information and Knowledge Management","author":"Robertson","year":"2004"},{"key":"2026040314280606700_ref051","first-page":"346","article-title":"Viewing term proximity from a different perspective","volume-title":"Advances in Information Retrieval (ECIR 2008), Springer LNCS 4956","author":"Song","year":"2008"},{"key":"2026040314280606700_ref052","doi-asserted-by":"crossref","first-page":"779","DOI":"10.1016\/S0306-4573(00)00015-7","article-title":"A probabilistic model of information retrieval: Development and comparative experiments. Part 1","author":"Sparck Jones","year":"2000","journal-title":"Information Processing and Management"},{"key":"2026040314280606700_ref053","doi-asserted-by":"crossref","first-page":"809","DOI":"10.1016\/S0306-4573(00)00016-9","article-title":"A probabilistic model of information retrieval: Development and comparative experiments. Part 2","author":"Sparck Jones","year":"2000","journal-title":"Information Processing and Management"},{"key":"2026040314280606700_ref054","first-page":"295","article-title":"An exploration of proximity measures in information retrieval","volume-title":"Proceedings of the 20th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Tao","year":"2007"},{"key":"2026040314280606700_ref055","doi-asserted-by":"crossref","DOI":"10.1145\/1183614.1183698","article-title":"Optimisation methods for ranking functions with multiple parameters","volume-title":"Fifteenth Conference on Information and Knowledge Management (ACM CIKM)","author":"Taylor","year":"2006"},{"key":"2026040314280606700_ref056","unstructured":"Terrier, \n           (n.d.), Homepage. http:\/\/ir.dcs.gla.ac.uk\/terrier."},{"key":"2026040314280606700_ref057","first-page":"12","article-title":"PF\/Tijah: Text search in an XML database system","volume-title":"Proceedings of the 2nd International Workshop on Open Source Information Retrieval (OSIR)","author":"van Os","year":"2006"},{"key":"2026040314280606700_ref058","author":"van Rijsbergen","year":"1979"},{"key":"2026040314280606700_ref059","doi-asserted-by":"crossref","first-page":"1","DOI":"10.6028\/NIST.SP.500-246","article-title":"Overview of the eighth text retrieval conference (TREC-8)","volume-title":"The Eighth Text Retrieval Conference (TREC-8), NIST Special Publication 500-246","author":"Voorhees","year":"2000"},{"key":"2026040314280606700_ref060","unstructured":"Wumpus\n            \n           (n.d.), Homepage. Wumpus. http:\/\/www.wumpus-search.org\/."},{"key":"2026040314280606700_ref061","unstructured":"Xapian\n            \n           (n.d.), Homepage. Xapian. http:\/\/xapian.org."},{"key":"2026040314280606700_ref062","article-title":"Microsoft Cambridge at TREC 2004: Web and HARD track","volume-title":"The Thirteenth Text Retrieval Conference (TREC 2004), NIST Special Publication 500-261","author":"Zaragoza","year":"2005"},{"key":"2026040314280606700_ref063","unstructured":"Zettair\n            \n           (n.d.), Homepage. Zettair. http:\/\/www.seg.rmit.edu.au\/zettair."}],"container-title":["Foundations and Trends\u00ae in Information Retrieval"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.emerald.com\/ftinr\/article-pdf\/4\/1-2\/1\/11486410\/1500000019en.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/www.emerald.com\/ftinr\/article-pdf\/4\/1-2\/1\/11486410\/1500000019en.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T14:32:00Z","timestamp":1777473120000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.emerald.com\/ftinr\/article\/4\/1-2\/1\/1326508\/The-Probabilistic-Relevance-Framework-BM25-and"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2009,9,17]]},"references-count":63,"journal-issue":{"issue":"1-2","published-print":{"date-parts":[[2009,9,17]]}},"URL":"https:\/\/doi.org\/10.1561\/1500000019","relation":{},"ISSN":["1554-0669","1554-0677"],"issn-type":[{"value":"1554-0669","type":"print"},{"value":"1554-0677","type":"electronic"}],"subject":[],"published":{"date-parts":[[2009,9,17]]}}}