Discussion of relevance and the needs of the users in information retrieval focuses on a deductive object-oriented approach and suggests eight inference rules for the deduction. Highlights include characteristics of a deductive object-oriented system, database and data modeling language, implementation, and user interface. (Contains 24…

Investigates document ranking methods in thesaurus-based Boolean information retrieval systems and proposes a new thesaurus-based ranking algorithm called the Extended Relevance algorithm. Performance comparisons are made between the Extended Relevance algorithm and previous thesaurus-based ranking algorithms. (Contains 20 references.) (LRW)

Presents a tested hypothesis raised by an earlier study and confirmed that query-surrogate documents chosen from within dense interdocument structures in scaled and visualized TREC Topic document data sets increased the number of relevant documents retrieved by 26-46%. Performance of information retrieval systems on data sets for which relevant…

Compares results of traditional Boolean searching with those of Freestyle, LEXIS/NEXIS's natural language application. Study found that though the Boolean searches had better results more often, neither method demonstrated superior performance for every query, suggesting that different queries demand different techniques. Concludes that further…

Discusses various search strategies and tools that can be used for searching on the Internet, including search engines and search directories; Boolean searching; metasearching; relevancy ranking; automatic phrase detection; backlinks; natural-language searching; clustering and cataloging information; image searching; customization and portals;…

Describes a study that examined the effectiveness of eight search engines for the World Wide Web. Calculated traditional information-retrieval measures of recall and precision at varying numbers of retrieved documents to use as the bases for statistical comparisons of retrieval effectiveness. Also examined the overlap between search engines.…

Discusses evidence-based medicine (EBM) and the need for systematic use of databases like MEDLINE with more sophisticated search strategies to optimize the retrieval of relevant papers. Describes an empirical study of hospital libraries that examined requests for information and search strategies using both structured and unstructured forms.…

Discusses similarity measures that are used in information retrieval to improve precision and recall ratios and presents a combined vector-based distance and angle measure to make similarity measurement more scientific and accurate. Suggests directions for future research. (LRW)

Develops a mathematical model for fitting the retrieval runs in the TREC-6 (Sixth Text REtrieval Conference) ad hoc retrieval exercise to explore how the depth of the pool of relevance judgments made for each run affects performance scores. Concurs that even on the basis of analyses of individual systems, the pool depth of 100 used by TREC is…

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…

Discussion of finding relevant documents in electronic document collections focuses on an evaluation of the Kea automatic keyphrase extraction algorithm which was developed by members of the New Zealand Digital Library Project. Results are based on evaluations by human assessors of the quality and appropriateness of Kea keyphrases. (Author/LRW)

Describes the MultiText information retrieval system that finds arbitrary passages of text, as opposed to complete documents, that are likely to be relevant. Discusses relevance ranking, term expansion, interactive user interface, and distributed searching used in experiments for TREC-6 (Text Retrieval Conference) in the ad hoc, high precision,…

Describes an experiment in automatic categorization, which is based on the vector model, widely used in information retrieval. Shows how the construction of the class patterns was carried out. Discusses the evaluation measures adopted and results obtained in the automatic categorization of a collection of documents in Spanish. Describes the manual…

According to signal detection theory (SDT), retrieval warnings may decrease false memory in the associative list paradigm either by inducing a conservative criterion shift or by decreasing the amount of evidence that critical theme words were studied. Fitting a SDT model to 12 existing datasets revealed suggestive evidence that warnings impact…