Information Retrieval List Digest 092 (November 20, 1991) URL = http://hegel.lib.ncsu.edu/stacks/serials/irld/irld-092 IRLIST Digest November 20, 1991 Volume VIII, Number 49 Issue 92 ********************************************************** II. QUERIES B. Requests for Information III. JOB ANNOUNCEMENTS 1. Assistant Professor, Cognitive and Neural Systems, Boston University IV. PROJECT WORK B. BIbliographies 1. IR-Related Dissertation Abstracts ********************************************************** II. QUERIES II.B.1. Fr: Paul F. Schulze Re: Request for Sources of Information I'm a new subscriber to this list, and am looking for sources of information. I have been given the task of planning for an Information/user support center with a user support help desk and help phone/hotline service. However, I have been asked to justify the investment before proceeding. The type of information that is wanted is: 1. Why should an organization have a Information Center/Help Desk operation? 2. What are the services a Information Center/Help Desk would provide? 3. Who in an organization will make use of such a service? 4. Where in the organization should it be located? 5. How should it be supported(free=costs borne centrally, or fee for service or?) 6. What are the implementation costs and staffing costs? 7. When is it appropriate to start such an operation? Does the need for it decrease? end? ie should there be a "sunset" time? 8. What types of financial analyses have been done to show that the Information Center/Help Desk pays for itself? 9. What has been the reaction of the MIS community to non-MIS supported Information Center/Help Desks? 10. Has the Information Center/Help Desk been a good place for users gaining experience with software? for user training to take place? 11. Has the Information Center/Help Desk been a productivity enhancer? Is it a manifestation of or a cause of the development of end-user computing? 12. Has it enhanced both the information literacy and computer literacy of other organizations? 13. Has the Information Center/Help Desk enhanced communication and the up/downloading of information between the mainframe/mini-environment and the PC environment? Does anyone have an address (email or regular mail) for the Help Desk Institute (Colorado Springs CO????)? If any of this list's subscribers have any information I could use, I would appreciate it. If there are any articles that you know of that I should read, please let me know. Please send information to me directly at PFS@NIHCU. Thanks Paul F Schulze PFS@NIHCU.BITNET National Institutes of Health "Now in 1962 on National Center for Research Resources on a 7090..." WW 8A-12 Bethesda, MD 20892 All opinions/statements are my own and definitely do not pertain to the views/policies/regulations/laws of the US Government. ********************************************************** III. JOB ANNOUNCEMENTS III.1. Fr: David Lukas Re: Assistant Professor, Cognitive and Neural Systems, Boston University Boston University seeks to hire a tenure track assistant professor starting in Fall 1992 for its graduate Department of Cognitive and Neural Systems. The Department offers an integrated curriculum offering the full range of psychological, neurobiological, and computational concepts, models, and methods in the fields of neural networks, computational neuroscience, parallel distributed processing, and biological information processing, in which Boston University is a leader. Candidates should have extensive analytic or computational research experience in modelling nonlinear neural networks, especially in one or more of the areas: learning, speech and language processing, adaptive pattern recognition, cognitive information processing, and adaptive sensory-motor control. Send a complete curriculum vitae and three letters of recommendation to Stephen Grossberg, Chairman, Search Committee, Department of Cognitive and Neural Systems, Room 240, 111 Cummington Street, Boston University, Boston, MA 02215, no later than January 1, 1992. Boston University is an Equal Opportunity/Affirmative Action employer. If you have questions or require further information, please reply to Carol Jefferson---caroly@cns.bu.edu. ********************************************************** IV. PROJECT WORK IV.B.1. Fr: Susanne M. Humphrey Re: Selected IR-Related Dissertation Abstract 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. AN University Microfilms Order Number ADG91-08846. AU JIMISON, HOLLY BRUGGE. TI A REPRESENTATION FOR GAINING INSIGHT INTO CLINICAL DECISION MODELS. IN Stanford University Ph.D. 1990, 205 pages. DE Health Sciences, General. Computer Science. Engineering, Biomedical. Information Science. AB It is often difficult to give an intuitive summary and explanation of a complicated decision problem. Yet, if a computer decision-support system is to be successful, it must provide the user with insight into the decision model and a justification for the resulting recommendations. Many clinical models can be characterized as weak, meaning that there is significant uncertainty in both the data input and inferences. This lack of determinism makes it especially difficult for system users to understand and have confidence in the models. This dissertation presents a representation for uncertainty and patient utilities that serves as a framework for graphical summary and computer-generated explanation of patient-specific clinical decision models. The implementation described is a computer decision aid designed to enhance the clinician/patient consultation process for patients with suspected angina (chest pain due to lack of blood flow to the heart muscle). In this particular domain, there is a great deal of uncertainty associated with a patient's prognosis given the clinical characteristics and test results. The generic angina model is represented as a Bayesian decision network, where the patient descriptors, probabilities, and utilities are treated as random variables. The initial distributions for these variables represent information on the population of patients with anginal symptoms, and the approach provides a method for efficiently tailoring the distributions to an individual patient. This framework also provides metrics for judging the importance of each variable in the model. The graphical interface uses this information to augment the display of a network representation of the model, and to generate a summary of the overall model uncertainty for a clinical user. Variables that are important for clinician/patient communication are highlighted in the graphical display of the network. The text version of the explanation is used to tailor a Hypermedia implementation of patient-education materials to an individual patient. The generated explanation emphasizes variables that: (1) Contribute significantly to a change in the treatment utilities between the generic and the patient-specific model; (2) Deviate from typical clinical presentations; (3) Remain sensitive to the treatment decision. These techniques serve to keep the explanation of the patient's decision model concise, allowing the communication with the patient to focus on the most important aspects of the treatment decision. AN University Microfilms Order Number ADGDX-91840. AU COLES, SUSAN. TI PERSONAL INFORMATION SYSTEMS: THE IMPLICATIONS OF JOB AND INDIVIDUAL DIFFERENCES FOR DESIGN. IN University of Technology, Loughborough (United Kingdom) Ph.D. 1990, 369 pages. DE Information Science. Business Administration, General. AB Available from UMI in association with The British Library. In an age where information has become a crucial commodity, accessing appropriate information quickly is essential to economic success. Developing ways of improving information retrieval is therefore of central concern to human factors and technologists alike. One aspect of information access relates to the ability of individual office workers to manage and retrieve their own information effectively, and this is what the present research addresses. Previous work in the area has been dominated by designing computer interfaces for the average user. This research investigates how people's needs might differ according to circumstance and examines a wider range of design possibilities. Specifically it sets out to relate retrieval problems (specific information retrieval rather than e.g. browsing or reminding) to job and individual (personality) differences within the general context of personal information management in offices using traditional technologies of paper, filing cabinets and desks. This is achieved by both extensive fieldwork and the use of simulated filing-retrieval systems in a controlled context. The work thus specifies conditions under which retrieval difficulties occur, and suggests how they might be causally operative. The findings permit the generation of hypotheses concerning: (1) Situations where retrieval problems are likely to be critical. (2) Novel avenues for improving performance, covering job design, changes to information management techniques (training and equipment design), and personnel selection. Techniques for predicting situations where poor retrieval could occur are put forward, and how methods of improving performance might be applied in offices explored. In the course of the research a number of methodological problems of pertinence both to the interpretation of present results and further developments in this area are identified and discussed. AN University Microfilms Order Number ADG91-09612. AU PALMQUIST, RUTH ANN. TI A STUDY OF WORD ASSOCIATIONS IN THE NATURAL LANGUAGE EXPRESSIONS OF INFORMATION NEEDS IN AN INFORMATION RETRIEVAL SETTING. IN Syracuse University Ph.D. 1990, 263 pages. DE Information Science. AB Two characteristics of a user's language behavior, vocabulary and structure (defined for this study as the statistical word associations found between vocabulary choices made), are used as indicators of a conceptual or knowledge structure. These characteristics are available in both the inquirer's statement or natural language expression (NLE) of his or her problematic situation and in the natural language portions of the document representations which are selected by the inquirer as useful to that context. This study sought to examine these language use characteristics, vocabulary and the resulting word associations, to determine whether they may be useful in discriminating relevant and non-relevant document representations. Using the automatic application of thesauri, user and document description vocabularies were simplified to provide a best possible case for the study. Word associations created by the order and proximity between co-occurring word pairs in both user NLEs and document descriptions with original and simplified vocabularies were determined. These word associations were compared against simple word frequencies for both original and simplified vocabularies. The findings suggest that the word associations which are identified in the user's NLE do significantly poorer at discriminating between the natural language or free-text portions of relevant and non-relevant document representations than do simple word frequencies. The simplification of vocabularies in both the user's NLE and the companion document representations created a "homogenizing" effect which essentially neutralized, overall, their ability to discriminate between a set of relevant and non-relevant document surrogates. These findings suggest that the user's conceptual structure, as determined using the vocabulary and statistical word associations in the NLE of his or her problematic context, is of limited value in the discrimination of relevant document descriptions from non-relevant ones. AN University Microfilms Order Number ADG91-09305. AU VON KEITZ, SAIEDEH ZAKARIA. TI AUTOMATED PATENT CLASSIFICATION FOR GERMAN PATENT DOCUMENTS. IN University of Saarland (Germany) Ph.D. 1989, 215 pages. DE Information Science. Library Science. AB This study deals with the problem of designing an automated system for classifying patent documents. First the efforts to apply automation to library classification schedules are reviewed. Since automation should provide better access to the body of technical knowledge, an automated patent system must be supported. To organize patent information, the International Patent Classification (IPC), which is accepted worldwide, is used. It is a one-dimensional hierarchic classification system. German patent documents are stored in a databank, PATDPA. PATDPA contains over 600,000 documents published since 1981. Because of the special structure of patent applications (claims and description) the information directly related to the invention and the information additional and supplementary to the invention can be helpful for assigning class numbers through an automated system. The German Patent Office makes available several aids to facilitate retrieval of information from classified documents. The Directory of Alphabetical Order Patent Descriptors and Index Terms is a list of the technical and common names of processes, machines, articles, composition of matter and other technological terms. This directory is used as a guide, rather than a precise locater. It should also be published in numerical order. Once the user learns the appropriate class with the approximate subclass of interest, then the directories in class number order can be helpful. These directories are used in the suggested automated classification system. To obtain the innovative information and the concept of the patent application, its full text should be indexed. The two keyword indexing techniques, KWIT (keyword in title) and KWOT (keyword out of text), are suggested for indexing the title and full text to extract the terms representing the concept of the patent application. Through the occurrence analysis and weighting of these terms and with the help of the directories, the appropriate class numbers can be assigned. During this process the directories can be continuously completed and updated. This system can also improve and correct the classification of already classified documents. The application of the International Patent Classification as an instrument for subject and class number search in the German patent databank PATDPA is illustrated. Suggestions are made for more effective retrieval in PATDPA. The above-mentioned processes and suggestions can bring the idea of thesaurus construction for each of the eight sections of the International Patent Classification into reality. The suggested system simplifies the processing of patent documents, avoids mistakes and results in a well classified collection of patent documents which is obviously fundamental to determining patentability of a patent application. A uniform classification facilitates worldwide search into the existence of patent rights. AN University Microfilms Order Number ADG91-06803. AU KRZYS, RICHARD ANDREW. TI ANALYSIS OF SHORES'S ENCYCLOPEDICS THEORY AND ITS APPLICATION TO A PROPOSED LATIN AMERICAN STUDIES ENCYCLOPEDIA OF RESEARCH. IN University of Pittsburgh Ed.D. 1990, 143 pages. DE Library Science. Education, General. AB The problem of this research project was to analyze an exploratory study that applied the Encyclopedics Theory to the design of a prototype for a Latin American Studies Encyclopedia of Research (LASER) for the purpose of making recommendations for a comprehensive LASER study. ********************************************************** 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.