Introduces a novel approach to unsupervised text summarization. Proposes an "information-centric" approach to evaluation, where the quality of summaries is judged not in terms of how well they match human-created summaries but in terms of how well they represent their source documents in information retrieval tasks such as document…

Analyzes how students' growing understanding of the topic and search experience were related to their choice of search tactics and terms while preparing a research proposal for a small empirical study. In addition, findings are used to test Vakkari's (2001) theory of task-based information retrieval. (Author/AEF)

Discusses previous research that examined the effects of different kinds of vocabulary control on retrieval effectiveness, and describes a study that addressed the effects of indexing depth on standard retrieval evaluation measures in the context of a typical commercial retrieval system. It is concluded that controlled vocabularies affect…

Investigates retrieval performance of controlled and uncontrolled subject representations as a function of retrieval expectation in the cystic fibrosis (CF) database. (Contains 22 references.) (KRN)

Describes the use of the tetrachoric correlation for experimental research and hypothesis testing of the efficiency of computerized information searching. Highlights include recall, relevance, and precision in information retrieval; use of a spreadsheet; examples; and limitations. (Author/LRW)

Describes the Belief Function Model for automatic indexing and ranking of documents which is based on a controlled vocabulary and on term frequencies in each document. Belief Function Theory is explained, and the Belief Function Model is compared to the Standard Vector Space Model. (17 references) (LRW)

Reports on the use of stemming for Slovene language documents and queries in free-text retrieval systems and demonstrates that an appropriate stemming algorithm results in an increase in retrieval effectiveness when compared with nonstemming processing. A comparison is made with stemming of English versions of the same documents and queries. (24…

Discusses general issues of computer and human understanding; contrasts three paradigms of information retrieval methodology, including statistical, deep semantic or natural language, and smart Boolean; describes CONIT, a knowledge-based intermediary retrieval assistance system; and examines system evaluation procedures, including a…

Synthesizes major findings from research on the magnitude of individual differences in information retrieval (IR) tasks and suggests implications for practice and design of IR systems and interfaces. Topics discussed include indexing consistency, relevance judgments, selection of subject headings, selection of search terms, and search retrievals.…

Presents a new measure of retrieval evaluation called corrected precision, which refines the traditional measures of recall and precision by taking into account differences among users in their selectivity when judging document relevance. The formula for calculating corrected precision is explained, and results from an empirical study of relevance…

An analysis of 100 online public access catalog (OPAC) search sessions explored the effectiveness of converting unsuccessful subject searches to keyword searches. It is suggested that either a system generated message to the user to try a keyword search, or an automatic conversion of zero-hit searches might be effective in improving search…

Describes prototype retrieval systems that use statistically based rank retrieval of records for text retrieval and bibliographic records: (1) the PRISE System at the National Institute of Standards and Technology; (2) the CITE System, for MEDLINE; (3) the Muscat System at the Scott Polar Research Institute; and (4) the News Retrieval Tool at the…

Presents the Semantic Vector Space Model, a text representation and searching technique based on the combination of Vector Space Model with heuristic syntax parsing and distributed representation of semantic case structures. In this model, both documents and queries are represented as semantic matrices, and retrieval is achieved by computing…

Compares three different approaches of user profile acquisition in a filtering system for customizing information acquisition on the World Wide Web. Explains the SIFTER system designed at Indiana University and discusses results of an analysis of variance that showed exclusive reliance on relevance feedback led to inferior filtering performance.…

Describes a session on information retrieval systems that planned to discuss relevance measures with Web-based information retrieval; retrieval system performance and evaluation; probabilistic independence of index terms; vector-based models; metalanguages and digital objects; how users assess the reliability, timeliness and bias of information;…