Information Retrieval List Digest 205 (March 21, 1994) URL = http://hegel.lib.ncsu.edu/stacks/serials/irld/irld-205 IRLIST Digest ISSN 1064-6965 March 21, 1994 Volume XI, Number 12 Issue 205 ********************************************************** I. NOTICES A. Meeting Announcements/Calls for Papers 1. The Clinical Information Station B. Publications Announcements 1. ARL Proceedings C. Miscellaneous 1. Summer School on Computer Linguistics II. QUERIES B. Requests for Information 1. Automatic Indexing/Frequency Analysis 2. Morphological Analysis and Stemming IV. PROJECT WORK C. Abstracts 1. IR-Related Dissertation Abstracts ********************************************************** I. NOTICES I.A.7. Fr: Theodore Morris Re: The Clinical Information Station Preliminary Announcement Please mark your calendar for a one day conference to be held in Cincinnati on June 23, 1994. The conference is entitled "The Clinical Information Station: Concept and Reality: The Point, Promise, Patient Record (and other Pieces), Path, Pilot, Problems, Progress, Plan, and Price. The Conference is sponsored by the University of Cincinnati Medical Center Libraries, the Greater Midwest Region of the National Network of Libraries of Medicine, the Friends of the National Library of Medicine, Merck, Inc., and other corporate sponsors. THE PROGRAM: The Point and the Promise: Why do we care and what is the vision; Colonel John Silva, M.D., Uniformed Services University of the Health Sciences ------------------------------------ The Patient Record/Database and other Pieces: What are the key components; Mark Frisse, M.D., Washington University School of Medicine ----------------------------------- The Path: What is the model and required infrastructure? Richard C. Howe, Ph.D., University of Cincinnati ----------------------------------- The Pilot Clinical Information Station: One person's view; J. Robert Beck, M.D., Baylor College of Medicine ----------------------------------- A Panel Discussion (Problems, Progress, Plan & Price) Moderator: Nancy M. Lorenzi, Ph.D. Panelists: John Silva, M.D.; Mark Frisse, M.D.; Richard Howe, Ph.D.; J. Robert Beck, M.D.; Hugh Hawkins, M.D. (University of Cincinnati); Mary Ellen Mahoney, M.S., R.R.A. (University of Cincinnati); Betsy Weiner, Ph.D., R.N. (University of Cincinnati) ------------------------------------- The registration fee is $100, which includes meals and refreshments. For further information or to receive a conference brochure and enrollment form, contact Diana Smith at: Voice: (513)558-5656 Fax: (513)558-2682 Internet: Diana.Smith@uc.edu ********** I.B.1. Fr: Association of Research Libraries Re: Proceedings Publication Scholarly Publishing on the Electronic Networks Proceedings of the Third Symposium November 13-15, 1993 ARL has published these proceedings of the meeting which was sponsored by the Association of American University Presses, the NSF, and the University of Virginia Library. The objective of this annual symposium (initiated in 1992) is to promote information-sharing and discussion among those interested in developing the potential of formal scholarly electronic publishing, with particular emphasis on not-for-profit models. Presentations discussed library models, electronic text issues, economic concerns such as cost recovery, and intellectual property laws. Scholarly Publishing on the Electronic Networks: Proceedings of the Third Symposium (February 1994) ISBN 0-918006-73-2. 184 pages, 8.5 x 11, pbk. (The Proceedings of the Second Conference are also available.) For more information contact Ann Okerson, Director, Office of Scientific and Academic Publishing. ann@cni.org. To order, contact ARL Publications, 202/296-2296 (phone), 202/872-0884 (fax), or osap@cni.org. ********** I.C.1. Fr: Nicolas Nicolov Re: Summer School on Computer Linguistics International Summer School "CONTEMPORARY TOPICS IN COMPUTATIONAL LINGUISTICS" 1 - 5 Sept 1994 Varna, Black Sea Coast, BULGARIA Sponsored by: European Association for Machine Translation and ECCAI-European Coordinating Committee for Artificial Intelligence Who should attend: The school is intended mainly for university students and young researchers, but will offer both introductory and advanced courses. Those who would like to extend their stay in Bulgaria can join the Bulgarian National seminar on Mathematical and Computational Linguistics which will take place from the 4th to the 10th of September 1994 at the same place. PRELIMINARY PROGRAMME: (three or four more courses will be added) Margaret King (ISSCO, University of Geneva, Switzerland) Evaluation of Machine Translation Systems Sergei Nirenburg (Carnegie-Mellon University, Pittsburgh, USA) Latest Developments in Machine Translation Pieter Seuren (University of Nijmegen, Netherlands) Introduction to Semantic Syntax Wolfgang Wahlster (DFKI, Saarbruecken, Germany) Multilingual Natural Language Interfaces Rodolfo Delmonte (University of Venice, Italy) Discourse Structure and Reference resolution Manfred Kudlek (University Hamburg, Germany) Models for Time, Tense and Aspect in Natural Languages Carlos Martin-Vide (Universidad Rovira i Virgili, Tarragona, Spain) Mathematical Linguistics: Its relevance for Computational Linguistics and Cognitive Science ORGANISERS: R.Mitkov Institute of Mathematics BULGARIA M.Zock LIMSI, Orsay FRANCE M.Kudlek University of Hamburg GERMANY N.Nikolov Incoma-TD Co, Ltd, Shumen BULGARIA FOR COMPLETE INFORMATION, CONTACT People wishing to participate should contact one of the people mentioned below. Ruslan Mitkov E-mail: ruslan@cslab.kaist.ac.kr Nicolas Nicolov E-mail: nicolas@aisb.edinburgh.ac.uk Nikolai Nikolov Tel: +359-54 56948 Fax: +359-54 56881 ********************************************************** II. QUERIES II.B.1. Fr: Y. Shum Re: Automatic Indexing/Frequency Analysis Hi there, One of the tasks in my project involves creating a software to extract keywords/key phrases from a text. At the moment, I have reached the stage where I could obtain the frequency of keywords of a document. I know that there are 101 ways of proceeding from here. Yet I'm not really sure of the proper algorithm to perform the frequency analysis and subsequently derive key phrases from them. I will appreciate and acknowledge any hints/suggestions of how to proceed from here. Thank you very very much. ********** II.B.2. Fr: Y. Shum Re: Morphological Analysis and Stemming Hi there, I'm doing a project which involves automatically generating indexes given a text. I have arrived at the stage where I need to reduce a word to its stem. I would like to know if there is any existing software system to which I could feed a word and get back its stem. Alternatively, I would like to learn of any good alogrithm to perform the stemming of words/morpholgical analysis of words. At the moment, my way of doing it is rather crude. I feed a word to the WordNet (an online lexicon), scan the result, and extract the stem from the result. It only works for reducing a plural noun to its singular, comparative (e.g., quicker) and superlative (e.g., quickest) adjective to its stem (e.g., quick), gerund verb (e.g., fighting), past tense verb (e.g., fought), past participle verb (e.g., shown) and singular verb (e.g., fights) to its original stem. And this method requires a lot of coding/hacking. Thank you very very much. ********************************************************** IV. PROJECT WORK IV.C.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 ADGNN-73929. AU BROWN, MARILYN A. TI THE INTEGRATION OF ONLINE BIBLIOGRAPHIC INFORMATION INTO STUDENTS' RESEARCH PROCESS. IN University of Toronto (Canada) Ed.D. 1992, 277 pages. SO DAI v53(12), SecA, pp4137. DE Education, General. Library Science. IS ISBN: 0-315-73929-0. AB The research study investigated three interrelated questions: What information search process do students conduct. What types of information and sources do students consult, use and integrate into research projects. What impact does online bibliographic information have on students' research process. One class of students enrolled in an Ontario Academic Credit World Issues geography course was studied. The study, a quasi ethnography, used two interactive data collection methods of key-informant interviews (students and teachers) and survey instruments (structured student interview schedule, written questionnaire, structured teacher interview schedule) and one non-interactive method of artifact collection (research journal, essay bibliography, student study guide, literature search results). No students could articulate a structured research model nor did they write a plan for their research project. Most students declared themselves to be procrastinators and all seemed resigned to their work habits. Most students equated selecting a topic to focusing a topic but focusing did not occur until they began to write under pressure of the due date. Locating and organizing information and returning to the library for more information occurred while writing. Books were the preferred source of information as they were perceived to be more specific and factual and they could be taken home. Newspapers articles were collected as part of the assignment but few students saw them as being sources of information for their essay. Eighteen students requested online searches, seven used some of the citations and four cited materials (magazines) in their bibliographies. Students did not use the search results because they had enough information from their books or they did not understand how to interpret and use the results. Students and teachers agreed that online searching would be useful at the beginning of a research project because it would save time, provide more current information and facilitate more specific research. The study concluded that online bibliographic searching will have little impact until students improve their information search skills, understand the complexity of the research process, improve organization and time management skills, and change their attitudes. AN University Microfilms Order Number ADG93-11009. AU HALBECK, PATRICIA GISELA. TI THE DEVELOPMENT OF A COMPUTER DATABASE TO SELECT PIANO REPERTOIRE FROM THE ROMANTIC ERA FOR INTERMEDIATE STUDENTS. IN The University of Oklahoma D.M.A. 1992, 189 pages. SO DAI v53(12), SecA, pp4244. DE Education, Music. Music. Education, Technology. AB This document develops a computer data base that aids teachers in the selection of piano repertoire from the Romantic era for intermediate level students. Repertoire by Burgmulller, Grieg, Gurlitt, Heller, Karganov, Kirchner, Mendelssohn, Schumann, and Tchaikovsky is included in the data base. Repertoire entries created for 100 pieces include the following information: title, composer, key, time signature, level of difficulty, mood, prominent rhythm pattern or note value, length, tempo, selected editions, and performance skills. Performance skills focus on technical and musical patterns such as scales, chords, double notes, articulation, textures, pedalling, ornaments, and dynamics. The user can access pieces by a single element of information such as key, or tempo, or by combinations of elements such as mood, tempo, and specific performance skills. The program uses IBM compatible materials and can be operated from a 3 1/2" high density floppy disk or loaded onto a hard disk for greater speed. The document includes a hard copy of all repertoire entries, user instructions for the program, and a listing of the data base program. The data base on computer disk is available from the author. AN University Microfilms Order Number ADG93-09124. AU BATY, JAMES BOBBITT, II. TI BUILDING THE ELECTRONIC SHOPPING MALL: A MULTIMEDIA ARCHITECTURE FOR ELECTRONIC COMMERCE. IN The University of Texas at Austin Ph.D. 1992, 79 pages. SO DAI v53(12), SecA, pp4113. DE Information Science. Computer Science. Business Administration, Marketing. AB This research proposes a design approach to Electronic Shopping Systems which addresses limitations in existing systems and suggests a data model for increased product differentiation and comparability, and enhanced navigation. Existing Electronic Shopping Systems offer very limited product information, navigation aids, or shopping assistance. These systems tend to use very traditional system design and emphasize text information with menu based navigation. Participating vendors are limited in their ability to differentiate their products and services from others', and customers are limited in their ability to compare product offerings. In these electronic markets both product differentiation and comparability are restricted. In spite of the advanced technology available, the electronic shopping experience is often worse than traditional physical shopping or mail order. This research utilizes multimedia to increase information content, frame-based representation to provide flexible data structures, and an object oriented approach to binding information presentation to data description to address the limitations in ESSs. A data model is proposed which provides for the description of individual store product hierarchies in a multiple vendor environment. A prototype is demonstrated which includes graphical browsing and search based navigation, with specific provision for product differentiation and comparability. It is asserted that the functionality provided by this approach can enable a shift in the vendor/customer dialectic from the perspective where differentiation and comparability are restricted, and in opposition, to one where vendors have increased ability to market their products in desired ways and customers will have increased ability to "shop and compare". Beyond the specified application, this architectural model may also be applied to other environments where the user must navigate multiple heterogenous data structures. AN University Microfilms Order Number ADGNN-73783. AU OLANDER, ANNA BIRGITTA. TI PERSONAL INFORMATION MANAGEMENT IN COMPUTER SCIENCE RESEARCH. IN University of Toronto (Canada) Ph.D. 1992, 227 pages. SO DAI v53(12), SecA, pp4114. DE Information Science. Computer Science. IS ISBN: 0-315-73783-2. AB Computer scientists' behaviour as they seek and use information in their research was studied in order to identify personal characteristics in information management. The central research questions were: (1) How does a computer scientist manage the information that he or she uses in research and teaching. (2) What implications do personal and environmental characteristics have for the way individual computer scientists handle scholarly information. (3) What are the implications of personal information management for the provision of scholarly information to researchers in computer science. Qualitative research methods were chosen as the most suitable for assessing factors critical for information-seeking behaviour. Nine computer science researchers in a Swedish university were studied. Each participant constituted a case. Information management was analyzed from the individual user's perspective. The study belongs to what Dervin and Nilan call the "alternative paradigm" of information needs and uses research (ARIST 1986). Interviewing was the main instrument for data collection. The most important findings are that an analysis of the participants' information behaviour has led to a typology, in matrix form, of different modes of information behaviour. This matrix combines explicit and implicit information needs and internal and external information seeking into four sets of criteria: explicit needs/external seeking, explicit needs/internal seeking, implicit needs/external seeking, and implicit needs/internal seeking. The participants' information management is focused on satisfying implicit information needs through internal information seeking. Information provision to researchers like the participants must support this kind of information behaviour. My study has also shown that the participants prefer colleagues over printed sources. Conference proceedings are the preferred printed sources. I have also found that the participants filter information input by exposing themselves only to the most relevant material. In order to adequately provide scholarly information to computer scientists like the participants, support for informal, internal information seeking and wide exposure to information should be developed. Methods for input filtering; evaluation support, and flexible access are required to facilitate personal information management geared to meet the implicit information needs that dominate in this user group. AN University Microfilms Order Number ADG93-11331. AU PHILLIPS, ROBERT LLOYD. TI THE RELATIONSHIP BETWEEN WORK ROLES AND INFORMATION SEEKING BEHAVIORS AMONG SELECTED PROTESTANT MINISTERS IN TARRANT COUNTY, TEXAS. IN University of North Texas Ph.D. 1992, 193 pages. SO DAI v53(12), SecA, pp4117. DE Library Science. Religion, Clergy. Education, Psychology. Information Science. AB Is there a relationship between work role and information seeking behavior. Three behaviors were examined: choice of information channel, choice of method for information retrieval, and choice of method for storing retrieved information. The Protestant clergy was selected as a profession with clearly identified work roles, including preacher and administrator. Questionnaires were mailed to 150 randomly selected ministers in Tarrant County, Texas. Sixty-four responded. Additionally, fifteen ministers selected at random were interviewed for additional data. The data collected through the questionnaires were analyzed using nonparametric statistical techniques. The research showed relationships between information seeking behaviors and both choice of information channel and choice of method for retrieving information. The research did not show a relationship between work role and choice of method for storing information. When choosing an information channel, the minister as preacher preferred a relatively few formal information channels, especially a personal book collection, a personal pamphlet file, and books specifically purchased for the sermon. In the administrator role, the minister was more likely to use a wide variety of informal information channels and only a few formal channels. These informal channels included another minister, a member of the congregation, or a staff colleague. A preferred formal channel for administration was printed information received from a denominational office. When choosing a method for retrieving information, the minister as administrator preferred those methods which provided direct, immediate access to the information channel, especially interpersonal contact or a telephone conversation. The minister as preacher preferred individualistic or impersonal means for accessing information such as reading a book. In addition to work role, the study also found a relationship between information seeking behaviors and years of experience, educational background, size of congregation served, denominational affiliation, and dollars spent to add materials to one's personal library. It could not show a relationship between age and the choice of information channels. AN University Microfilms Order Number ADG93-10245. AU QUEEN, STEPHEN EUGENE. TI INFORMATION DISPLAY AND SEARCH STRATEGIES. IN The University of Connecticut Ph.D. 1992, 74 pages. SO DAI v53(12), SecB, pp6599. DE Psychology, Industrial. Computer Science. AB The pervasiveness of powerful microcomputers and presentation software has enabled unsophisticated users to easily create and modify the configuration and format (graphic and alphanumeric) of information intended to influence decision makers. Research shows that decision makers will search for information in multiattribute decision problems by alternative or by attribute depending on how information is configured. Only recently has the issue of display format been investigated within the context of display configuration. The present study examined these issues in greater detail using a new computerized process tracing technique. Thirty-three university undergraduate students made 60 decisions comprised of four alternatives and four attributes. Each decision problem was displayed using a bar chart or alphanumeric chart. Each chart was grouped by alternative, or by attribute, or was presented in a neutral fashion. Each decision problem contained attribute sets of high or low importance weight dispersion. A computerized information board, based on Mouselab (Johnson, Payne, Schkade, & Bettman, 1989) was developed to monitor subjects' information acquisitions, response times, and choices. Subjects could reveal individual bars or numbers in a sequential fashion using a "mouse". This study found that display configuration can have a significant impact on a decision maker's search behavior by simply arranging information in different ways on a single display. It was also found that subjects had lower relative accuracy when information was grouped by attribute, which may have implications for consumer publications that typically group information by attribute. No differences in search behavior or relative accuracy were found between bar charts and alphanumeric charts, which contradicts previous research. Bar charts and alphanumeric charts, however, were presented in a sequential fashion. The theoretical implications of displaying information sequentially versus simultaneously is explored. Finally, this study confirmed previous research, finding that subjects search for less information, are more selective about the information they do search for, search for information primarily by attribute, and have higher relative accuracy for decision problems with high importance weight dispersion. ********************************************************** 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 calur@uccmvsa.ucop.edu or calur@uccmvsa.bitnet Nancy Gusack ncgur@uccmvsa.ucop.edu or nancy.gusack@ucop.edu Mary Engle meeur@uccmvsa.ucop.edu or mary.engle@ucop.edu The IRLIST Archives is now set up for anonymous FTP, as well as via the LISTSERV. Using anonymous FTP via the host dla.ucop.edu, the files will be found in the directory pub/irl, stored in subdirectories by year (e.g., /pub/irl/1993). Using LISTSERV, 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 LOGYYMM, where YY is the year and MM is the numeric month in which the issue was mailed, to LISTSERV@UCCVMA (Bitnet) or LISTSERV@UCCVMA.UCOP.EDU. You will receive the issues for the entire month you have requested. 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.