Discusses software design issues for Okapi, an information retrieval system that incorporates both search engine and user interface and supports weighted searching, relevance feedback, and query expansion. The basic search system, adjacency searching, and moving toward a distributed system are discussed. (Author/LRW)

Examines general issues in conducting information retrieval research. Topics include the Okapi information retrieval system and its probabilistic model; the Cranfield projects, concerning recall and precision; the SMART project with its vector-space model; evaluation methodology, including laboratory evaluation of interactive systems; and…

Proposes a theoretical framework called NLPIR that integrates natural language processing (NLP) into information retrieval (IR) based on the assumption that there exists representation distance between queries and documents. Discusses problems in traditional keyword-based IR, including relevance, and describes some existing NLP techniques.…

Discusses genetic algorithms in information retrieval, especially for relevance feedback, and evaluates the efficacy of a genetic algorithm with various order-based fitness functions for relevance feedback in a test database. Compares results with the Ide dec-hi method, one of the best traditional methods. (Contains 56 references.) (Author/LRW)

Presents the application of a knowledge visualization tool, VxInsight[R], to enable domain analysis for science and technology management. Uses data mining from sources of bibliographic information to define subsets of relevant information and discusses citation mapping, text mapping, and journal mapping. (Author/LRW)

Describes an experiment that investigated how relevance feedback could be used to build and adjust profiles to improve filtering system performance. Explains the use of the SIFTER (Smart Information Filtering Technology for Electronic Resources) filtering system by graduate students and discusses results that established the importance of context…

Provides background on the problems of traditional text retrieval systems and describes STATUS/IQ, an advanced text retrieval system that incorporates a natural language front-end and an advanced relevance ranking facility. The principles, capabilities, and benefits of the system are discussed, and an example of a STATUS/IQ session is presented…

Discussion of retrieval overlap and relevant overlap focuses on an analysis of data from an earlier study on overlap among document representations. This new analysis investigated the odds that an item was judged relevant when it was retrieved by two search fields as opposed to being retrieved by only one field. (Contains 17 references.) (LRW)

Describes the development of a method for the automatic identification of sublanguage vocabulary words as they occur in abstracts. Highlights include research relating to sublanguages and their vocabulary; domain terms; evaluation criteria, including recall and precision; and implications for natural language processing and information retrieval.…

Classifies information-retrieval applications into three groups depending on the correspondence between a user's request and the queries posed to the document base. Argues that the mapping of requests, on a semantic level, to formalized queries, on a lexical level, determines the range of retrieval effectiveness. (Author/LRW)

Describes SWEAR (Systematic Weighting and Ranking), a powers-of-two algorithm that can be used for searching the World Wide Web or any large database that automatically creates discrete, well-defined result sets and displays them in decreasing order of likely relevance. Also discusses fuzzy sets. (Author/LRW)

Discussion of natural language processing focuses on term weighting in information retrieval. Presents a new weighting method that depends on low frequency terms, called negative weighted inverse verb frequency, and discusses case frames, word similarity, similarity measurement, and recall and precision improvement. (LRW)

Evaluates the retrieval performance of an algorithm that automatically categorizes medical documents, which consists in assigning an International Code of Disease (ICD) based on well-known information retrieval techniques. Reports on experimental results that tested precision using a database of over 20,000 medical documents. (Author/LRW)

Reports on an experiment that examined whether thesaurus terms, related to query in a specified semantic way (synonyms, narrower terms, related terms, or broader terms) could be identified as having a more positive impact on retrieval effectiveness when added to a query through automatic query expansion. (Contains 54 references.) (Author/LRW)

Compares empirical results from three experiments using Text Retrieval Conference (TREC) data and search topics that involved three different user interfaces. Results show that marking Boolean queries on text, which encourages browsing, and hypertext interfaces to text retrieval systems can benefit recall and can also benefit novice users.…