Information Retrieval List Digest 093 (November 25, 1991) URL = http://hegel.lib.ncsu.edu/stacks/serials/irld/irld-093 IRLIST Digest November 25, 1991 Volume VIII, Number 50 Issue 93 ********************************************************** I. NOTICES B. Publications Announcements 1. NSF/Argonne SciDB BlitbBd: Multidisciplinary Forum on Scientific Database Research II. QUERIES A. Questions and Answers 1. Answer to Lou Rosenfeld: /Verity B. Requests for Information 1. Query: Hierarchic Similarity Measures IV. PROJECT WORK B. Bibliographies 1. IR-Related Dissertation Abstracts ********************************************************** I. NOTICES I.B.1. Fr: editor@eid.anl.gov (Lucien Russell) Re: NSF/Argonne SciDB BlltbBd Multidisciplinary Forum on Scientific Database Research Argonne National Laboratory, in support of the National Science Foundation's Initiative on Scientific Database Research, announces a new, multi-disciplinary electronic Forum, the Scientific Database Bulletin Board (SCID3B), to spur collaborative efforts among scientists and database researchers. The Scientific Database Bulletin Board was created to help overcome the numerous barriers that impede the development of interdisciplinary research efforts. By providing an easily accessible information service, the Bulletin Board provides a forum that will bring researchers with scientific data management problems together with those dedicated to solving them. Argonne National Laboratory with its broad spectrum of research efforts in the biological, physical, earth and information sciences provides a microcosm of the scientific community, and is thus an ideal home for such a service. In addition, researchers in the social sciences who manage massive statistical databases are a target membership of the Forum as well. This initial call for participation opens the Forum by posing the fundamental question "What are the major issues facing scientific data management, and what data management services should be provided?". Participation in the debate is solicited from all interested persons in the Natural, Social and Computer/Information Sciences. From the responses gathered a taxonomy of issues will be created, and special interest Sub-Forums will be established in the SCID3B. In addition, the SCID3B will contain directories of collections of scientific data, lists of upcoming conferences and workshops, and over time, a posting of researchers interested in collaborating on scientific or statistical database projects. Future plans include both a software distribution and data visualization service. Most important, the SCID3B will publish a quarterly Newsletter, and will recruit an Editorial Board to publish a Journal of Scientific Database Research. To encourage the widest participation of academic, government, and commercial organization the SCID3B will be available free through Internet. All interested parties are urged to contact the Editor, Lucian Russell at the electronic mail address EDITOR@SCID3B.EID.ANL.GOV with a White Paper of up to two pages on the issues. Researchers interested in serving on an Editorial Board for a Sub-Forum on a particular topic should indicate this interest. E-mail access exists now, but it is expected that the Bulletin Board will initiate full service 1/2/1992. The editor requests that this message be re-distributed by any recipient to any and all potential interested parties world-wide, and a copy posted on office Bulletin Boards, especially those of organizations that do not widely use electronic mail. Please note: All material in the SCID3B Foundation under Agreement IRI-9121721. Any opinion, findings and conclusions or recommendations expressed are those of the authors and do not necessarily reflect the views of the National Science Foundation. Lucian Russell, Editor in Chief Scientific Database Bulletin Board (SCID3B) ********************************************************** II. QUERIES II.A.1. Fr: Edward A. Fox Re: IR-Related Dissertation Abstracts The following are citations selected by title and abstract as being related to Information Retrieval (IR), resulting from a computer search, using BRS Information Technologies, of the Dissertation Abstracts Online database produced by University Microfilms International (UMI). Included are UMI order number, title, author, degree, year, institution; number of pages, one or more Dissertation Abstracts International (DAI) subject descriptors chosen by the author, and abstract. Unless otherwise specified, paper or microform copies of dissertations may be ordered from University Microfilms International, Dissertation Copies, Post Office Box 1764, Ann Arbor, MI 48106; telephone for U.S. (except Michigan, Hawaii, Alaska): 1-800-521-3042, for Canada: 1-800-268-6090. Price lists and other ordering and shipping information are in the introduction to the published DAI. An alternate source for copies is sometimes provided. Dissertation titles and abstracts contained here are published with permission of University Microfilms International, publishers of Dissertation Abstracts International (copyright by University Microfilms International), and may not be reproduced without their prior permission. From a population of ninety-three centers of Latin American studies twenty-eight were randomly selected from three strata, each with a different size faculty ranging from fifty or fewer than twenty members. A total of two-hundred twenty-four questionnaires were mailed to twenty-eight directors of the centers for distribution to individuals who by teaching or research were heavily committed to Latin American studies. Fifty questionnaires, or a 20% response rate, was obtained. The study demonstrated the relevance of the Shores's Encyclopedics theory to the design of the LASER; however, it also indicated the need for modifying Shores's principles in accordance with recent tendencies and technologies and with data derived from Latin American studies. The prototype identified the purposes of the LASER to create, summarize, identify, and stimulate research in topics dealing with Latin American studies. Nine major recommendations resulted from the LASER study as follows: (1) Funding for a comprehensive LASER study should be obtained. (2) The Encyclopedics Theory should be brought up-to-date in light of changes in media and technology. (3) A time-table allowing adequate time for return of responses should be followed. (4) A rigorous study should be conducted of the classification scheme used within the questionnaire to determine its future use in the LASER itself. (5) More than one version of the questionnaire should be devised for different audiences of the LASER. (6) The sample size should be increased to include at least 60% of the survey subjects listed in the Bray/Greenleaf Directory. (7) Techniques for ensuring a representative sample, e.g. f and t Tests should be utilized to test samples. (8) Researchers should utilize personal contacts of colleagues to achieve a larger response from the survey audience. (9) The exploratory study should be followed by a comprehensive study incorporating the recommendations generated from the exploratory study. AN University Microfilms Order Number ADG91-07290. AU TEACH, BEVERLY. TI THE INDIANA UNIVERSITY AUDIO-VISUAL CENTER FILM/VIDEO LIBRARY: AN ANALYSIS OF USE AND USERS BY MAJOR SUBJECT AND USER GROUPS. IN Indiana University Ph.D. 1990, 231 pages. DE Library Science. Information Science. AB This study is an in-depth, use-centered evaluation of the Indiana University Film/Video Library. Circulation records spanning two fiscal years were examined with regard to general patterns of use, use by subject and client group, and characteristics of Indiana University faculty use and users. Descriptive statistics, cumulative frequency and Bradford distributions, proportional use statistics, and nonparametric measures of correlation were used to analyze the data. Observations on the relationship between proportional use statistics and the level of analysis were made. Cumulative frequency distributions of print circulations and title refusals corresponded to the positively skewed patterns common to use studies of print materials. Contrary to common findings in studies of obsolescence, for all subjects except one, both the proportion of titles requested and mean requests per title for titles produced in the 1960s were as great as or greater than titles produced in the 1970s or 1980s. Titles acquired in the 1980s within one year of production were requested more frequently than titles acquired even two years after production, but no more than titles acquired in the 1960s. Findings confirm that there is often a delay in the circulation of recently acquired film/video materials and point out the effect regular weeding of noncirculating titles can have on circulation patterns over time. Overlap of requests by client groups was highest between and among secondary schools, Indiana University, and other higher education clients, with substantial repeat use of titles from year to year. Client groups generally requested titles within recommended grade-levels but, except for Indiana University and other higher education clients, they exhibited marked differences in specific subjects most requested. Within Indiana University, faculty in different disciplines exhibited different subject request patterns. Faculty of rank Assistant Professor and above accounted for over two-thirds of all requests. The amount of direct instructional support provided in terms of total credit hours and course enrollment was highest in lower level courses; however, courses using film/video had above average enrollment at all levels. The average number of film/video titles shown per course was higher in freshman level courses but held steady across all other levels. AN University Microfilms Order Number ADG91-09538. AU BROWN, TONY J. TI VISUAL DISPLAY HIGHLIGHTING AND INFORMATION EXTRACTION. IN Tufts University Ph.D. 1990, 120 pages. DE Psychology, Experimental. Psychology, Industrial. Computer Science. AB This research examined the influence of computer-generated highlighting on search performance and information extraction. The tasks used to examine information extraction were a detection task, a localization task, and an identification task. For each of these three tasks there were twelve highlighting techniques that were presented among displays of varying sizes (i.e., 3, 12, 24 items). The study tested twenty-five subjects and they viewed a total of 30,480 screens. The results suggested three components of processing that played a role in extracting highlighted information. An initial scan of the entire display, which required little attention, but allowed the detection of color or closure. Localization, which required more detailed attention, produced significantly longer response times than detection and significantly shorter response times than identification. This result suggested that location information is not preattentively available and does require focused attention. All highlighting techniques tested were detected and localized more quickly and accurately than the non-highlighted condition. The identification task required the greatest amount of attention and suggested a final stage of analysis that allowed recognition of highlighted information. There were cases where "pop out" was found with the identification task, which provided support for some theories and contradicted others. During all three tasks the items highlighted by color or closure were not influenced by changes in set size and produced significantly faster response times than the other highlighting techniques. There were also strong brightness-contrast effects that significantly influenced search. Apparent depth produced no significant differences in this study. One highlighting technique (blinking) produced an interaction between task type and set size, which suggested the possibility of independent stages of processing. There was also partial support for the notion that some techniques may draw attention away from the item being highlighted, while others may allow the highlighted item to "pop out" at the viewer. Practitioners may find this paradigm and the results useful in helping to evaluate highlighting techniques for computer software displays. ********************************************************** IRLIST Digest is distributed from the University of California, Division of Library Automation, 300 Lakeside Drive, Oakland, CA. 94612-3550. Send subscription requests to: LISTSERV@UCCVMA.BITNET Send submissions to IRLIST to: IR-L@UCCVMA.BITNET Editorial Staff: Clifford Lynch lynch@postgres.berkeley.edu or calur@uccmvsa.bitnet Nancy Gusack ncgur@uccmvsa.bitnet Mary Engle engle@cmsa.berkeley.edu or meeur@uccmvsa.bitnet The IRLIST Archives will be set up for anonymous FTP, and the address will be announced in future issues. To access back issues presently, send the message INDEX IR-L to LISTSERV@UCCVMA.BITNET. To get a specific issue listed in the Index, send the message GET IR-L LOG ***, where *** is the month and day on which the issue was mailed, to LISTSERV@UCCVMA.BITNET. These files are not to be sold or used for commercial purposes. Contact Nancy Gusack or Mary Engle for more information on IRLIST. The opinions expressed in IRLIST do not represent those of the editors or the University of California. Authors assume full responsibility for the contents of their submissions to IRLIST.