Presents a probabilistic technique to retrieve passages from texts having a large size or heterogeneous semantic content. Results of experiments comparing the probabilistic technique to one based on a text segmentation algorithm revealed that the passage size affects passage retrieval performance; text organization and query generality may have an…

Discusses Internet domain naming practices and indicates which Web search engines can effectively search on domain names. Explains top-level domain (TLD) (ex. http://www.access.gpo.gov - ".gov" = top level; ".gpo" = second level; ".access" = third level; and "www" = fourth level domain). Outlines seven new…

Given technology's potentially negative effects on creativity, educators must examine the Internet's capabilities and best uses in the main office, school, and classroom. To locate superior Web sites, teachers should identify sponsors' affiliations, examine content, scrutinize ratings and awards, check listed references, and gather multiple…

Describes the building of a list of constructed-impact factors (CIF) for biomedical journals not included in the 1996 editions of the "Journal Citation Reports." The online retrieval from the host DIMDI of the data needed for impact-factor calculation is described. The top 100 (of 338 titles, ranked according to their CIFs) are…

Describes the development and application of visualization techniques for users to access and explore information in digital libraries effectively and intuitively. Salient semantic structures and citation patterns are extracted from several collections of documents using Latent Semantic Indexing and Pathfinder Network Scaling. Author cocitation…

Compares five World Wide Web search engines for precision on the first 20 results returned for 15 queries, adding weight for ranking effectiveness. Discusses methods to lessen evaluator bias, evaluation criteria, definition of relevance, experimental design, the structure of queries, and future work. (Author/LRW)

Compares the performance of the Target and Freestyle commercial search engines. Discusses traditional performance measures such as precision and recall, analytic models of performance, document rankings, query difficulty and correlates with other performance characteristics, and future research needs. (LRW)

Describes a relevance-feedback mechanism for content-based image retrieval that evaluates the feature distributions of the images judged relevant by the user and updates both the similarity measure and the query to accurately represent the user's information needs. (Author/LRW)

Discusses the need for the design of new user interfaces for digital libraries and describes SenseMaker, a user-centered interface for information retrieval in a heterogeneous digital library. Topics include iterative bundling of articles and citations; evaluation results; increasing fluidity between browsing and searching; and filtering.…

Defines the notion of serial clustering of words in text, and explores the value of such clustering as an indicator of a word's bearing content. This approach is flexible in the sense that it is context-sensitive; a numerical approach may also be of value in assigning weights to terms in requests. Experimental support is obtained from natural text…

Uses a simple model of an information retrieval (IR) technology to examine performance measures for IR systems. Shows that, if a user is employing a system optimally, then reasonable performance measures will tend to be relatively robust, or insensitive, to small variations in assumptions about the user's behavior. (AEF)

Describes the fifth Dublin Core (DC Workshop) DC is a small set of simple, descriptive data elements intended to aid in the discovery of Internet resources. Covers metadata issues, "DC Lite," current international projects utilizing the Core, and issues yet to be resolved. (AEF)

Discusses research conducted by the National Historical Publications and Records Commission on electronic record keeping (tools and techniques to manage electronic records produced in an office environment) and on the best practices for storing, describing, and providing access to all electronic records of long-term value. (Author/LRW)

Describes a study that was conducted to identify, describe, and analyze the document representations of Web sites revealed in the hit list of subject searches conducted by four search engines: AltaVista, infoseek Ultra, Lycos, and Open Text Index. Problems are discussed, and implications for Web page design are considered. (LRW)

Explains how to train users in effective Web searching. Discusses challenges of teaching Web information retrieval; a framework for information searching; choosing the right search tools for users; the seven-step lesson planning process; tips for delivering group Internet training; and things that help people work faster and smarter on the Web.…