Information Retrieval List Digest 103 (February 27, 1992) URL = http://hegel.lib.ncsu.edu/stacks/serials/irld/irld-103 IRLIST Digest February 27, 1992 Volume IX, Number 7 Issue 103 ********************************************************** I. NOTICES A. Meeting Announcements/Calls for Papers 1. Neural Networks for Learning, Recognition and Control 2. IASSIST Conference: Data, Networks, and Cooperation: Data, Networks, and Cooperation: Linking Resources in a Distributed World C. Miscellaneous 1. Usefulness Measure IV. PROJECT WORK B. Bibliographies 1. IR-Related Dissertation Abstracts ********************************************************** I. NOTICES I.A.1. Fr: dlukas@park.bu.edu Re: Neural Networks for Learning, Recognition and Control CALL FOR PAPERS International Conference on NEURAL NETWORKS FOR LEARNING, RECOGNITION, AND CONTROL May 14-16, 1992 Gail A. Carpenter and Stephen Grossberg CONFIRMED SPEAKERS: May 14: R. Shiffrin, R. Ratcliff, D. Rumelhart. May 15: M. Mishkin, L. Squire, S. Grossberg, T. Berger, M. Bear, G. Carpenter, A. Waxman, T. Caudell. May 16: G. Cybenko, E. Sontag, R. Brockett, B. Peterson, D. Bullock, J. Albus, K. Narendra, R. Pap. CONTRIBUTED PAPERS: A featured 3-hour poster session on neural network research related to learning, recognition, and control will be held on May 15, 1992. Attendees who wish to present a poster should submit three copies of an abstract (one single-space page), post-marked by March 1, 1992, for refereeing. Include a cover letter giving the name, address, and telephone number of the corresponding author. Mail to: Poster Session, Neural Networks Conference, Wang Institute of Boston University, 72 Tyng Road, Tyngsboro, MA 01879. Authors will be informed of abstract acceptance by March 31, 1992. A book of lecture and poster abstracts will be given to attendees at the conference. For information about registration and the two neural network tutorial courses being taught on May 9-14, call (508) 649-9731 (x255) or request a meeting brochure in writing when submitting your abstract. ********** I.A.2. Fr: Anne L. Cooper UW-CDE Data Libr. 608/262-6323 or 238-5575 Re: IASSIST Conference, Madison, WI, May 26-29, 1992 DATA, NETWORKS, AND COOPERATION: LINKING RESOURCES IN A DISTRIBUTED WORLD The 18th annual conference of the International Association for Social Science Information Service and Technology (IASSIST) will be held at the Concourse Hotel in MADISON, WISCONSIN, U.S.A. from TUESDAY, MAY 26 through FRIDAY, MAY 29, 1992. The conference theme expresses IASSIST members' concern for managing and sharing computer-readable data during a time of increasing demand coupled with decreasing fiscal resources. The conference program features workshops, contributed papers, roundtable discussions, and poster sessions reflecting international viewpoints on these concerns. IASSIST brings together individuals engaged in the acquisition, processing, maintenance, and distribution of computer-readable text and numeric social science data. Founded in 1974, the membership includes data archivists, librarians, information specialists, social scientists, researchers, and government agency administrators from around the world. IASSIST '92 CONFERENCE AND WORKSHOPS Conference WORKSHOPS, Tuesday, May 26, 1992 9:00 Introduction to Unix, Coordinator: Jim Jacobs, UC - San Diego or Introduction to Data Archives and DataLibraries Coordinator: Jean Stratford, UC - Davis 2-5:00 Navigating Unix: Resources and Applications Coordinator: Juri Stratford, UC - Davis or Software Access to 1990 Census CD-ROMS, Coordinator: Jack Beresford, Bureau of the Census ---- Preliminary Summary of CONFERENCE Sessions Wednesday, May 27, 1992 a.m. Concurrent Sessions Text and Tools: Resources for the Humanities Chair: John Price-Wilkins, University of Michigan or Vietnam Data in the National Archives Chair: Peggy Adams, National Archives & Records Administration p.m. Concurrent Sessions Archiving Electronic Records, Ch: Karsten Rasmussen,Danish Data Archives or Quantitative Analysis & Stylistics: When the Letters Meet the Numbers Chair: Laura Bartolo, Kent State University Thursday, May 28, 1992 a.m. Concurrent Sessions Standards, Chair: TBA or Migrating Between Systems, Ch: Pat Hildebrand,University of Pennsylvania p.m. Poster Sessions Friday, May 29, 1992 a.m. Plenary Session a.m. Concurrent Sessions Integrating the Use of Numeric Data in the Academic Environment Chair: Wendy Treadwell, University of Minnesota or Data from Wisconsin, a Panel of Principal Investigators Speakers: Jim Sweet, Larry Bumpass, Robert M. Hauser,Erik Wright, University of Wisconsin or MARC Meets NASA, Chair: Laine Ruus, University of Toronto p.m. Concurrent Sessions Panel of Professional Associations Chairs: Tom Brown, National Archives & Records Administration Diane Geraci, SUNY Binghamton or Issues in Health Data Research and Management, Chair: TBA Entertaining activities have been planned for the evenings. Registration fees received by April 20, 1992 are as follows: $175 Conference and Workshop; $125 Conference only $ 75 Workshop only or One-Day Conferance Attendance Program Committee Chair: Ilona Einowski, Data Archivist, UC Data, University of California - Berkeley 2538 Channing Way, Berkeley, CA 94720 U.S.A., phone: 510-642-6571 CENSUS85@UCBCMSA.BITNET TO RECEIVE FINAL PROGRAM AND REGISTRATION MATERIALS, PLEASE CONTACT THE LOCAL ARRANGEMENTS COMMITTEE: LEW@WISCSSC.BITNET or LEW@SSC.WISC.EDU Cindy Lew, Associate Data Librarian, Center for Demography & Ecology, University of Wisconsin - Madison, Madison WI 53706 phone:608-262-9827 fax:608-262-8400 GUY@MACC.WISC.EDU Laura Guy, Data Librarian, Data and Program Library Service, University of Wisconsin - Madison, Madison WI 53706 COOPER@SSC.WISC.EDU Anne Lightfoot Cooper, Data Librarian, Center for Demography & Ecology, University of Wisconsin - Madison, Madison WI 53706 ********** I.C.1. Fr: Peter Schauble Re: Usefulness Measure: Source available The usefulness measure is an effectiveness measure to compare two retrieval methods as described in H.P. Frei, P. Schauble (1991), Determining the Effectiveness of Retrieval Algorithms, Information Processing & Management, Vol. 27, Nos 2 & 3, pp. 153-164. The main advantages of the usefulness measure are the following. First, it comes with an error probability that expresses how reliable the derived usefulness value is. Second, the usefulness measure is based on preferences rather than on a dichotomy of relevant and non-relevant documents. Finally, partial relevant assessments are sufficient to determine the usefulness of a retrieval method. Thus, determining the usefulness is feasible even when the document collection is extremely large and dynamic. Anyone interested in the usefulness measure should send a request to schauble@inf.ethz.ch. I will send you a UNIX shell archive containing the necessary C programs, a makefile, an example, and a README file. Peter Schauble ********************************************************** IV. PROJECT WORK IV.B.1. Fr: Susanne M. Humphrey Re: Selected 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. AN University Microfilms Order Number ADG91-15674. AU SAWYER, JEANNE CLIFFORD. TI A REFERENCE AND PLANNING MODEL FOR LIBRARY ONLINE PUBLIC ACCESS CATALOGS. IN The University of North Carolina at Chapel Hill Ph.D. 1990, 177 pages. DE Computer Science. Library Science. Information Science. AB This dissertation explores the problem of how library catalogs can be connected or combined, providing library users with access to materials beyond the local library collection while maintaining local autonomy and ease of implementation. The first step in solving the problem was to identify the fundamental ways library systems can be connected, and to determine the characteristics of each. The Reference Model for Online Public Access Catalogs (OPAC Model) does this by categorizing the basic architectures into three models: centralized, distributed and stand-alone. The reference model provides a way to classify a system according to its architecture and identifies the basic characteristics that a system must have. The distributed model (DLN Model) is explored in depth because it is the least well understood of the models and because the characteristics of distributed systems must be standardized if such systems are to be connected effectively. Whereas the OPAC Model defines the system architectures, the Library Systems Architecture Planning Model (LSAP Model) provides a tool for choosing among them. The system characteristics defined in the reference model are included to meet real-world needs, such as providing access to another library's holdings or preserving local autonomy. The LSAP Model follows from the Reference Model by making explicit the connection between a set of system characteristics and a set of environmental characteristics. The concepts included in the Reference Model are new and untested, especially for the distributed architecture. Therefore a case study of the Triangle Research Libraries Network's system was included in the dissertation specifically to demonstrate that: (a) the reference model can be implemented, and (b) the implementation is reasonable and is an appropriate choice for that environment. Verifying the LSAP Model was then necessary to demonstrate that the planning model works, i.e., that the Model accurately reflects expert judgements of appropriate choice of system architecture. In addition, verification of the LSAP Model further validates the Reference Model since the LSAP Model is built from the architectures delineated in the Reference Model. If those architectures were inappropriately defined, the LSAP Model could not work properly. AN University Microfilms Order Number ADG91-16638. AU MOHAGEG, MICHAEL FREDRICK. TI THE INFLUENCE OF HYPERTEXT LINKING STRUCTURES AND TASK-RELATED VARIABLES ON INFORMATION RETRIEVAL TASKS. IN Virginia Polytechnic Institute and State University Ph.D. 1990 243 pages. DE Engineering, Industrial. Computer Science. Information Science. Engineering, System Science. AB Hypertext is a method of online information management and/or presentation where textual documents are parsed (modularized) into many nodes and inter-connected using machine-supported links. These systems have become increasingly popular in numerous applications. Unfortunately, few empirical investigations have been conducted concerning the usability and utility of hypertext, and the effusive claims made by many hypertext enthusiasts are largely unsubstantiated. This study investigated several usability issues relating to hypertext within the context of an information retrieval application. Of particular interest were system linking structures consisting of linear, hierarchical, network, and combination hierarchical/network configurations. These commonly used hypertext linking structures were imposed on a text-intensive geography database (GEO). GEO contains 187 nodes discussing a variety of topics concerning the countries of North Africa. In addition to the linking structures, the task variables of number of required links (to reach the answer) and task type were studied. Task type refers to expert programmers' judgements as to whether a task is best suited to a hierarchical or network linking structure. The approach was to create a set of information retrieval (IR) tasks with specific characteristics (as determined by number of required links and task type), and to study the performance of each linking structure in completing these tasks. The intention was to identify the task situations under which each linking structure excels. Results indicate that hierarchical linking structures perform quite well for most IR tasks and perform significantly better than network linking. The combination condition performs no worse than hierarchical, yet, with the exception of task completion times, provides no consistent advantages over the hierarchical structure. Hence, it is concluded that, for novice users of a system, the performance advantages resulting from the inclusion of network links (in isolation or in combination with hierarchical) are not commensurate with the associated costs of creating such links. Ultimately, results are aimed at a better understanding of hypertext systems, their performance, and more judicious applications of these systems. AN University Microfilms Order Number ADG91-17174. AU EZE, MOSES O. J. TI SIGNATURE ANALYSIS TECHNIQUE FOR DIAGNOSING FAILURES IN MECHANICAL SYSTEMS. IN Colorado State University Ph.D. 1990, 123 pages. DE Engineering, Mechanical. Engineering, Industrial. AB There is widespread interest in diagnostic systems that can predict impending mechanical system failures. This study focuses on the use of acoustic data to determine when and why a machine is malfunctioning. The need to minimize unpredicted failures, reduce maintenance costs, and increase machine availability, especially when complex processes characterized by either hostile environments or inaccessible areas are involved, has led to a growing interest in the development of diagnostic/prognostic capabilities. Some techniques dealing with signal analysis have been satisfactorily implemented, but they are limited to special applications. Humans have traditionally employed sounds emitted from moving mechanical parts as diagnostic tools to predict and detect problems in machines. For example, automobile mechanics typically use sound as an important sensory input to diagnose the state of a car's engine. This type of analysis is dependent upon the skill level of the listener. On one hand, the sound one hears varies from person to person. For instance, some people cannot hear certain frequency ranges. Even the most skilled listener is susceptible to a head cold, which jeopardizes the ability to detect sounds. On the other hand, each listener must hear many machines, both defective and nondefective, to establish a basis for deciding if a machine is defective. By recording the acoustic sounds from a given machine and using a computer to perform a portion of the data analysis, the failures can be better predicted and consequently unscheduled machine downtime can be minimized. This study investigates the use of a manufacturing tool that supplements the human ear's ability to detect mechanical problems in a complex mechanical device. In other words, the research focuses on the detection of mechanical malfunctions through the employment of sound signals emitted from the failing device. The detection system uses a previously developed database of analyzed signals to characterize new and unknown signals. The signal parameters and characteristics generated through the analysis are used as input to the computerized signal database to search for a matching signal or the closest signal if one exists. A technique for dealing with situations for which no match exists in the database is also examined. Examples of the use of the developed failure-detection procedure are also given. Laboratory results indicate that the system performs well when provided with a good library of previously analyzed signals. AN University Microfilms Order Number ADG91-15084. AU DICKINSON, HOLLY JEAN. TI DERIVING A METHOD FOR EVALUATING THE USE OF GEOGRAPHIC INFORMATION IN DECISION-MAKING. IN State University of New York at Buffalo Ph.D. 1990, 176 pages. DE Geography. Computer Science. AB The research presented in this dissertation involves establishing the value of geographic information and its analysis in decision making. The discussion is focussed on the use of a Geographic Information System (GIS) in a decision-making organization. A literature search was performed to discover methods used in Economics, Management Science, and Information Science to establish the value of information. It is concluded that prior to establishing value, it is first necessary to improve our understanding of how geographic information is actually used. However, to support empirical observations of use, there is a need for a more structured format than descriptive case studies. A modeling technique, capable of revealing where geographic information is critical in a decision-making process and the costs and benefits associated with that use, is discussed. Specific characteristics of complex decision-making tasks are used as criteria in examining the applicability of various modeling techniques to this research. After a discussion of various techniques, Petri Nets are chosen. The ability of petri nets to represent geographic information use in complex decision-making tasks is shown through a case study in a forestry organization. The use of petri nets to attach and measure costs and benefits along each step of the process is presented at a conceptual level. The specific objective of this research is to demonstrate (through an in-depth case study) that the use of geographic information and its analysis can be modeled in sufficient detail to permit the identification of costs and benefits attached to all or part of the decision-making process. ********************************************************** 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.