The principle that, for optimal retrieval, documents should be ranked in order of the probability of relevance is discussed. (Author/KP)

Poses questions concerning methods of using specificity information in a weighting scheme. (JB)

Explores the possibility of combining the two principles of ranking according to degree of relevance, and ranking according to probability of relevance, for information retrieval systems, but concludes that, while neither is adequate alone, no single, all-embracing ranking principle can be constructed to replace the two. (Author/VT)

Examines statistical techniques for exploiting relevance information to weight search terms. These techniques are presented as a natural extension of weighting methods using information about the distribution of index terms in documents in general. (Author)

A model is proposed to explain the retrieval characteristics of an information retrieval system. (Author)

Presents adaptation of a probabilistic theoretical model previously used in relevance feedback for use in automatic indexing of documents (in the sense of imitating) human indexers. Methods for model application are proposed, independence assumptions used in the model are interpreted, and the probability of a dependence model is discussed.…

This two-part article combines a comprehensive account of a probabilistic model of retrieval with new systematic experiments on TREC (Text Retrieval Conferences) Program material. Part 1 covers the foundations and the model development for document collection and relevance data, along with the test apparatus. Data and results tables for both parts…

A comprehensive account of a probabilistic model of retrieval with new systematic experiments on TREC (Text Retrieval Conferences) Program material. Part 2 covers the further development of the model, with testing, and briefly considers other environment conditions and tasks, model training, concluding with comparisons with other approaches and an…

A Bayesian argument is used to suggest modifications to the Robertson and Jones relevance weighting formula to accommodate the addition to the query of terms taken from the relevant documents identified during the search. (Author)

Discusses term weighting formulae and their use for selecting new terms to enhance a search statement and for weighting the terms for retrieval purposes once selected. The Swets model of information retrieval system performance is described, an approach to term selection is presented, and future research is suggested. (five references) (LRW)

Reviews the development of research and evaluation in information retrieval. Highlights include the Cranfield projects concerning recall and precision, the Medlars experiment that focused on relevance, TREC (Text Retrieval Conference), evaluation versus understanding, interface versus functionality, document length, and terms and user interaction.…

Discusses the use of text retrieval methods based on probabilistic models with Chinese language material, which are modeled on the Okapi information retrieval system. Topics include system architecture, test collections, weighting functions, and algorithms. (Author/LRW)

The increasing complexity of evaluating information retrieval (IR) systems is discussed in terms of contributing factors and research methods. Examples are provided from experiments with weighted searching on a front-end system, information-seeking behavior and online catalogs, and the OKAPI experimental retrieval system. (21 references) (LAE)

Describes use of the Okapi system, an experimental text retrieval system at City University, London, on TREC (Text Retrieval Conference) test collections to investigate probabilistic models and weighting functions; automatic ad hoc experiments; and routing and filtering for query optimization. Discusses interactive system evaluation. (LRW)

Discusses the problem of determining an adequate sample size for an information retrieval experiment comparing two systems on separate samples of requests. The application of statistical methods to information retrieval experiments is discussed, the Mann-Whitney U Test is used for determining minimum sample sizes, and variables and distributions…