Relevance weighting of search terms

By Stephen E. Robertson, Karen Spärck Jones. 1976

Subfields: Information Retrieval & Search

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Retrieval & Search

Citations in the corpus

Inferring probability of relevance using the method of logistic regression (1994, background): “attribute IDF, which is usually logged, was first suggested by Karen Sparck-Jones [ 4 ].”
Modeling and solving term mismatch for full-text retrieval (2012, background): “These attempts include Smeaton (1997) and Voorhees (1999), and are typically characterized by their mixed performance or ineffectiveness.”
A formal study of information retrieval heuristics (2004, methodology): “Many different retrieval models have been proposed and tested, including vector space models [13, 12, 10], probabilistic models[7, 16, 15, 3, 6, 5], and logic-based models[17, 19, 2].”
Supervised and Traditional Term Weighting Methods for Automatic Text Categorization (2009, unknown): “RSJ (see [18]) is also known as the relevance weighting from the classical probabilistic model for IR.”
Adaptation of machine translation for multilingual information retrieval in the medical domain (2014, background): “where t ranges over the terms in q, k1, k3, and b are model parameters, tftd is the frequency of t in d, tftq is the frequency of t in q, Ld is the length of d, Lave is the average document length, and w(1) is the Robertson and Spärck Jones [129] weight of t in”
Naive (Bayes) at Forty: The Independence Assumption in Information Retrieval (1998, methodology): “Its use in the form of Equation 11 was promoted by Robertson and Sparck Jones in a paper [41] that did much to clarify and unify a number of related and partially ad hoc applications of naive Bayes dating back to Maron [37].”
Learning to Reweight Terms with Distributed Representations (2015, methodology): “BM25 is actually more directly connected to term recall weight, as the above original BM25 model is an extension of the BIM [16].”
The Simplest Thing That Can Possibly Work: Pseudo-Relevance Feedback Using Text Classification (2019, background): “The idea of treating document ranking as a binary classification problem, to distinguish relevant from non-relevant documents, has a long history, dating back to the binary independence retrieval (BIR) model of Robertson and Spark Jones [19].”
BizTalk Server 2000 Business Process Orchestration. (2001, unknown): “[21] Methods developed”
Essential Pages (2009, background): “Web-search is typically inﬂuenced by several metrics [1]: { C } content relevance [2] derived from documents’ anchor text [1], title and headings, word frequency and proximity [3], ﬁle, directory, and domain names, and other more sophisticated forms of content analysis.”
A comparison of event models for naive bayes text classification (1998, methodology): “The approach has been used for text classification by numerous people (Robertson and Sparck-Jones 1976; Lewis 1992; Kalt and Croft 1996; Larkey and Croft 1996; Koller and Sahami 1997; Sahami 1996).”