Describes an approach that combines concepts from information retrieval, fuzzy set theory, and genetic programing to improve weighted Boolean query formulation via relevance feedback. Highlights include background on information retrieval systems; genetic algorithms; subproblem formulation; and preliminary results based on a testbed. (Contains 12…

Describes an information retrieval methodology based on relevance weighting of search terms in pairs of documents. The vector space model of information retrieval is reviewed, the Bayesian model of text retrieval is discussed, and an automatic implementation of weight estimation using the Medline database is described. (16 references) (LRW)

Discusses the history of information science and its understanding of and approach to relevance; explains human categorization theory and applies it to the idea of relevance; shows that information science research is flawed because it fails to adequately explain experience; and suggests implications for future research. (17 references) (LRW)

To investigate the relationship between the retrieval mechanism and the level of question elaboration, this study divided 100 questions from the cystic fibrosis database into five conceptual categories based on their semantic representations. Two retrieval methods were chosen to investigate potential differences in outcomes across conceptual…

Examines the transition from traditional single database retrieval to the emerging networked resources retrieval, and presents research and experimental solutions developed at the University of California, Berkeley. Highlights include recall and relevance, network navigation, vocabulary, and extending retrieval records. (LRW)

Examines the statistical distribution of relevance judgments by reviewing and comparing the results of several studies over the last three decades. A characteristic distribution is examined, and a hypothesis is offered to explain it based on human judging characteristics and the nature of relevance experimentation. (Contains 13 references.) (LRW)

Reports preliminary results of a study that investigated the nature of feedback during 40 uses of user-intermediary-information retrieval system interaction through the use of microanalysis. Types of feedback based on human judgments include magnitude feedback, relevance feedback, term relevance feedback, tactical review feedback, and terminology…

Reports preliminary results of a research project that investigated the effectiveness of a new technique for information retrieval interfaces called colinked descriptors, based on the concept of cocited references in the field of bibliometrics. Examples of colinked descriptors from the "Thesaurus of ERIC Descriptors" are presented.…

Introduces navigation as a central concept for information retrieval and for information systems in general. The evolution of information systems is described; navigability as an aid to improve relevance is discussed; and a system under development that attempts to facilitate navigation for information retrieval is described. (31 references) (LRW)

Presents the Semantic Distance Model (SDM) of Relevance Assessment, a cognitive model of the relationship between semantic distance and relevance assessment. Discusses premises of the model such as the subjective nature of information and the metaphor of semantic distance. Empirical results illustrate the effects of semantic distance and semantic…

This session abstract discusses the history, philosophy, and theories of knowledge management to better understand its social and organizational potentials and limitations. Topics include determinacy of sense, information retrieval, and the Data Retrieval Model versus the Document Retrieval Model; discussions about knowledge; and surplus…

Uses a reliability framework to examine the role that relevance judgments play in quantitative information retrieval (IR) performance measurement with an eye toward issues introduced by Internet IR. Argues that a theoretical basis for reliable measurement is possible if the relevance yardstick is reconsidered in light of recent research and if the…

Reports on a continuing investigation of intertextual networks. Describes how intertextual networks can be modeled as directed graphs and extends this to matrix representations. Discusses topological index values of these networks and speculates how topological index values might be used in the estimation of retrieval values in information…

In assessing information retrieval systems, it is important to compare the number of retrieved relevant items to the total number of relevant items. A variant of the statistical "capture-recapture" method can be used in large collections to estimate total number of relevant documents, providing the model supporting such an analysis can…

Findings from three separate studies of relevance judgments by 44 initial online search users were examined. Results show that, for users conducting their initial search on a particular information problem, the number of items judged "partially" relevant, not item judged "highly" relevant, were positively correlated with important changes in…