Proposes that topic representations should consist of excerpts from domain thesaurus, generated by spreading activation technique. Algorithm for generating excerpts is outlined and exemplified, and problem of assessing resemblance between two excerpts is discussed. Whether existing thesauri are adequate for this purpose is questioned, and ideas…

Two articles discuss the clustering of composite representations in the Cystic Fibrosis Document Collection from the National Library of Medicine's MEDLINE file. Clustering is evaluated as a function of the exhaustivity of composite representations based on Medical Subject Headings (MeSH) and citation indexes, and evaluation of retrieval…

Describes a new text analysis approach to automate the indexing process, i.e., an automated keywording system designed to behave as a human expert indexer, and explains the rule base of the approach, which consists of lists of insertion and deletion rules generated by subject matter experts. (32 references) (Author/SD)

Describes research and evaluation on the Information Mapping method of structured analysis and writing of training and reference documents to illustrate how a flaw in research design can cause "little or no effect" results. Information retrieval experiments, learning experiments, and field tests are examined for problems, including size…

Discussion of the solicitation and measurement of relevance judgments focuses on a study of faculty and doctoral students at the University of Michigan that examined how users' judgments of document representations changed as more information about the documents was revealed to them. The use of magnitude estimation techniques is also discussed.…

Describes a collaborative, reuse hypertext system, called MUCH (Many Using and Creating Hypertext). The novel browsing and retrieval characteristics of the system are based on a semantic net connecting documents, thesauri, and discussions. Word patterns are documented for reuse, creating and reorganizing network traversal according to user…

Introduces the multicharacter trie as an index structure that can improve the time needed for retrieving full-text materials stored on CD-ROMs. The advantages of this structure compared to other structures are described, and experimental results comparing it to the widely used B+ tree and other structures used for full-text retrieval are…

The implementation and evaluation of a computer-based kanji finder developed on a Macintosh using HyperCard is reported. In a comparison study, the electronic kanji finder resulted in more effective retrieval than traditional retrieval using reference books. (18 references) (Author/LB)

Describes a low-cost, upgradeable, CD-ROM network and gives a step-by-step description of the installation process, including inventorying assets, learning about network components, selecting a network, installing adapter boards, attaching cabling, installing software, testing, configuring the server, installing search software, installing a menu,…

Discussion of online searching focuses on the first three of seven stages of searching: (1) the reference interview, including user needs, the search request, and the user's expectations; (2) the tactical overview, including institutional control and cost and time factors; and (3) database selection considerations, including relevance,…

Describes current indexing systems, documents the shift from manual to automated indexing methods, and offers suggestions for writers and editors on the critical importance of careful wording and controlled vocabularies to ensure effective access to technical and scientific articles. Suggests teaching strategies for helping students control…

This article continues the description of DFT (Document, Find, Theseus), an online documentation system that provides computer-managed on-demand printing of software manuals as well as the interactive retrieval of reference passages. Document boundaries in the hypertext database are discussed, search vocabulary complexities are described, and text…

Examined the use of the TREC Information Retrieval Test Collection for evaluating visual information retrieval interfaces (VIRI). Found that TREC data sets were suitable as development tools for VIRIs and traditional information retrieval systems and that the computational cost of organizing data into useful visual fields were high and related to…

Summarizes information that will be used to describe microform sets on the University of California, Irvine's World Wide Web site for microform resources, including title, publisher, price, contents, index/bibliography, reviewed in, location/holdings, and Webpac site. Discusses staff/network resources, HTML/UNIX editors, and indexing/search…

Potential problems for college and university computing functions posed by arrival of the year 2000 (Y2K) are discussed, including arithmetic calculations and sorting functions based on two-digit year dates, embedding of two-digit dates in archival data, system coordination for data exchange, unique number generation, and leap year calculations. A…