Describes the architecture and functioning of a system developed specifically for text processing applications--such as information retrieval--that can benefit from structural comparisons between elements of text, such as queries and abstracts. The general ways in which the system performs matches and the ways in which this objective influenced…

Presents a model that places Boolean database queries into conjunctive normal form, thereby allowing probabilistic ranking of documents and the incorporation of relevance feedback. Experimental results compare the performance of a sequential learning probabilistic retrieval model with the proposed integrated Boolean probabilistic model and a fuzzy…

This paper reports on a transaction log analysis of 51,473 queries from 18,113 users of Excite, a major World Wide Web search engine. Approximately 2,500 of these queries were from the use of relevance feedback. Given the high level of research activity and historical success of relevance feedback in assisting users in locating relevant…

Reports experiments with a term weighting model incorporating relevance information in which it is assumed that index terms are distributed dependently. It is argued that, if high recall searches are required, relevance feedback based on the modified dependence model may be superior to the Boolean search. (Author/VT)

Discusses the inability of the standard Boolean logic model of information retrieval to deal effectively with the inherent fallibility of retrieval decisions. Recent advances in information retrieval research are reviewed, and their practical potential for overcoming the deficiencies of the Boolean model is examined. (45 references) (Author/CLB)

A theoretical framework used to derive a weighting mechanism for ranking output documents in Boolean retrieval systems is described in detail. A simple illustrative example is included, and the implications of such an approach are discussed. (64 references) (Author/CLB)

Computers, though they can help educational institutions manage their facilities efficiently, are no substitute for good judgment. Before computerizing, institutions should systematize their records and evaluate their information needs; they should take care not to acquire unnecessary systems or become over-reliant on computers. (MCG)

Discussion of user-relevance judgments in information retrieval focuses on findings from four separate studies of relevance judgments by 55 users conducting their initial online search on a particular information problem. Highlights include a theoretical framework, models of relevance, and implications for information-retrieval system design and…

Reports on axiomatic approach to systems design that has emanated from documentation work in field of chemistry and closely related disciplines. It is based on five axioms: definability, order, sufficient degree of order, representational predictability, and representational fidelity. Mandatory indexing is discussed; a glossary is appended. (106…

Describes rough sets theory and discusses the advantages it offers for information retrieval, including the implicit inclusion of Boolean logic, term weighting, ranked retrieval output, and relevance feedback. Rough set formalism is compared to Boolean, vector, and fuzzy models of information retrieval and a small scale evaluation of rough sets is…

Discusses the probabilistic nature of information retrieval, and describes a front end system designed to permit searchers to attach probabilistically interpreted weights to query terms, which could be adapted for conventional information retrieval systems without requiring radical changes in current indexing procedures. (37 references) (CLB)

Describes the three subsystems of an information retrieval system (document descriptions, queries, and matching algorithms) and argues that the interdependency of these subsystems requires adaptation for the system to perform when any component changes. An algorithm for redescribing documents, in response to changes in queries and retrieval rules,…

Describes a new Boolean retrieval environment in which an extended soft Boolean logic is used to automatically construct queries from original natural language formulations provided by users. Experimental results that compare the retrieval effectiveness of this method to conventional Boolean and vector processing are discussed. (27 references)…

Discusses the motivation for integrating information retrieval and database management systems, and proposes a probabilistic retrieval model in which records in a file may be composed of attributes (formatted data items) and descriptors (content indicators). The details and resolutions of difficulties involved in integrating such systems are…

An online retrieval test was conducted using an experimental online catalog that provided several different subject searching strategies. Analyses of test data suggest that users expected alphabetical searches to provide lists of assigned subject headings conceptually related to their topics. System designers can now respond to their expectations.…