Information Retrieval List Digest 247 (March 13, 1995) URL = http://hegel.lib.ncsu.edu/stacks/serials/irld/irld-247 IRLIST Digest ISSN 1064-6965 March 13, 1995 Volume XII, Number 10 Issue 247 ********************************************************** I. QUERIES 1. Available keyword generators II. JOBS 1. UC Dublin: Information Studies, Lectureship III. NOTICES A. Publications 1. ToC: Information Processing & Management, 31:3 B. Meetings 1. ICML '95 2. Electronic Access to Government Information IV. PROJECTS A. Abstracts 1. IR-L Related Dissertation Abstracts ********************************************************** I. QUERIES I.1. Fr: Cammie Donaldson Re: Available keywords generators I am looking for commercially available, public domain, shareware, or otherwise obtainable software for generating keywords/phrases from a collection of documents. Documents can be in text (ASCII) format although tools dealing with special file formats (e.g., word processing) are of interest. I am not interested in something that produces a simple inverted index, but instead something that uses linguistic or statistical processing to produce keywords/phrases for individual documents in a collection. THANKS very much for any pointers you can give on where I should look for this info, or products that you know of, etc. Cameron M.G. Donaldson Software Productivity Solutions 122 4th Avenue Indialantic FL 32903 407/984-3370 cmd@sps.com ********************************************************** II. JOBS II.1. Fr: Dr. Mary A. Burke Re: UC Dublin, Lectureship, Information Studies @ U. C. Dublin UNIVERSITY COLLEGE DUBLIN Department of Library and Information Studies One Permanent Post (Ref: 22/95) Closing Date: Thursday 30th March 1995. Applications are invited for the above full-time academic post. As candidates should be capable of contributing strongly to the teaching of Information Studies in the BA Degree, an appreciation of the humanistic nature of Information Studies is essential. A willingness to teach and supervise research students on post-graduate programmes in Library and Information Studies is also desirable. The successful applicant should have a research track record which is commensurate with age and experience and the motivation to continue this research. The appointment will be made at the level of either Assistant Lecturer or College Lecturer. The current salary scales are: Assistant Lecturer: IR 13,513 - IR 21,809 College Lecturer: IR 21,065 - IR 34,148 Prior to application, further information (including application procedure) should be obtained from the Personnel Office, University College Dublin, Belfield, Dublin 4, Ireland (quoting above reference number). Telephone enquiries: (353 1) 706 1436 and (353 1) 706 1508 - direct lines. Fax: (353 1) 269 2472. UCD is an Equal Opportunities Employer ================================================= Post of Assistant / College Lecturer Department of Library and Information Studies THE DEPARTMENT: DepLIS currently has five full time academic staff, a secretary, many adjunct and tutorial staff, and about one hundred students. Programmes offered by the Department lead to a Bachelor of Social Science Degree, a Diploma in Library and Information Studies, a degree of Master of Library and Information Studies, a Master of Arts degree in Information Studies, or a Doctorate of Philosophy. Information Studies has recently been introduced in the Bachelor of Arts Degree, and will be offered from the start of the 1995/96 academic year. RESOURCES: The College Library is designed to meet most needs for printed resources. The College Audio-Visual Centre is also available for departmental use. Computing needs are met by a departmental PC based Novell Local Area Network, the extensive network of College mainframes and LAN servers, and by CD-ROM and online services and other Internet resources. THE JOB: The appointee will be required to contribute to the teaching, research and administrative activities of the Department. He / she will be required to teach approximately ten course units per annum, including a range of courses in Information Studies in the BA Degree. The syllabus for this Programme is enclosed as an addendum to this document. This does limit the appointee's areas of specialism within Information Studies, but it does impose a need for the successful candidate to be capable of teaching and researching Information Studies from a humanities perspective. He / she may also offer courses on the post-graduate programmes (DLIS, MLIS, MA) in Library and Information Studies. The Department will consider candidates with any specialism, and it may be possible to accommodate special interests by rearranging teaching within the Department. The appointee will also be expected to participate actively in research, and to supervise post-graduate student theses. These may include minor theses for the MLIS and MA programmes, and major theses for the MPhil and PhD Degrees. The lecturer's individual research may be independent of external funding, or it may involve the development of proposals for, and the management and execution of, EU funded projects. The appointee will be expect to undertake various administrative responsibilities as designated by the Head of Department. These are likely to include student counselling, co-ordination of specific academic programmes, membership of Departmental, Faculty, University or External Committees, and other miscellaneous activities. INTERVIEWS AND SEMINARS: It is hoped to hold interviews in the week of 2 - 5 May 1995. Candidates may be invited to make seminar presentations on the day preceding the interviews. FOR MORE INFORMATION, CONTACT: Dr. Mary A. Burke Head of Department Department of Library and Information Studies (DepLIS) University College Dublin (UCD) Belfield Dublin 4 Ireland Tel. +353 1 7067799; Fax. +353 1 7061161 Internet Email Address: MABURKE@OLLAMH.UCD.IE ********************************************************** III. NOTICES III.A.1. Fr: Lauren Levey Re: ToC: Information Processing & Management, 31:3 Volume 31, Number 3 1995 Special Issue: The Second Text Retrieval Conference (TREC-2) CONTENTS Donna Harman 259 The Second Text Retrieval Conference (TREC-2) Donna Harman 271 Overview of the Second Text Retrieval Conference (TREC-2) Karen Sparck Jones 289 Reflections on TREC Chris Buckley 313 Automatic Routing and Retrieval James Allan Using SMART: TREC-2 Gerard Salton James P. Callan 325 TREC and TIPSTER Experiments W. Bruce Croft INQUERY John Broglio S.E. Robertson 343 Large Test Collection Experiments S. Walker on an Operational, Interactive M.M. Hancock-Beaulieu System: Okapi at TREC Justin Zobel 359 Efficient Retrieval of Partial Alistair Moffat Documents Ross Wilkinson Ron Sacks-Davis Matt Mettler 377 TREC Routing Experiments with the Fritz Nordby TRW/Paracel Fast Data Finder David A. Evans 383 CLARIT-TREC Experiments Robert G. Lefferts Tomek Strzalkowski 395 Natural Language Information Retrieval William R. Caid 417 Learned Vector-Space Models for Susan T. Dumais Document Retrieval Stephen I. Gallant N.J. Belkin 429 Combining the Evidence of P. Kantor Multiple Query Representations E.A. Fox for Information Retrieval J.A. Shaw BOOK REVIEWS Thomas H. Martin 447 Telecommunications in Transition: Policies and Technologies in the European Community by C. Steinfeld, J.M. Bauer, and L.Caby (Eds.) Ray R. Larson 448 Using Subject Headings for Online Retrieval: Theory, Practice, and Potential by K.M. Drabenstott and D. Vizine-Goetz Dale H. Bent 449 Interactive Computer Systems: Videotex and Multimedia by Antone F. Alber B.A. Shuman 449 The Myth of the Electronic Library: Librarianship and Social Change in America by W.F. Birdsall Fabio Crestani 450 Neural Networks for Knowledge Representation and Inference by D.S. Lavine and M. Aparicio IV (Eds.) ********** III.B.1. Fr: Carlos Martin Vide Re: ICML'95 II INTERNATIONAL CONFERENCE ON MATHEMATICAL LINGUISTICS ICML'95 Tarragona (Catalonia, Spain) November 2-4, 1995 Organized by Rovira i Virgili University at Tarragona Research Group in Mathematical Linguistics and Language Engineering (GRLMC) With the financial support of Direccio General de Recerca (Generalitat de Catalunya) CALL FOR PAPERS ICML'95 intends to become an open and broad scope forum for the presentation and discussion of current high quality research on mathematical models of natural language. The first conference (ICML'93) took place in Tarragona on March 30-31, 1993. A derived volume has been published: C. Martin-Vide (ed.), Current Issues in Mathematical Linguistics. North-Holland, Amsterdam, 1994. Topics will be grouped into five sections: 1. Foundations and methodological considerations. 2. Mathematical models for syntax. 3. Mathematical models for semantics. 4. Applications in natural language processing. 5. Varia (phonology, computational complexity, learning, quantitative methods, etc.). The language of the conference is English. LECTURES: They can be invited plenary 1-hour lectures and free possibly parallel 30 minutes lectures. The travel and accommodation of invited speakers will be paid by the Organizing Committee. The distribution into both categories will take into account the relative value of each paper and the budgetary resources. Interested people are invited to submit papers: By surface mail, or Electronically, as either an unformatted (plain text) ASCII file or as a LaTeX file. They should send one copy of a camera-ready 2-columns abstract of not more than 2 not-numbered pages. Also, they must indicate the section into which their paper fits. Submissions will be anonymously refereed, and the list of accepted lectures (invited as well as free) will be included into future announcements. Authors will be notified of acceptance as soon as possible. TIMETABLE: Deadline for the submission of a paper (2-pages abstract): September 30, 1995. The programme will be sent to all pre-registered participants on October 1995. PUBLICATIONS: The book of abstracts will be available at the beginning of the conference. The Organizing Committee expects that a volume containing a refereed selection of submitted papers will be published after the conference. FOR COMPLETE INFORMATION: Carlos Martin-Vide Apartado de Correos 32.077 08080 Barcelona Spain E-mail: cmv@fll.urv.es Fax: 34-77-55.95.97 ********** III.B.2. Fr: Carol Ritzenthaler Re: Electronic Access to Government Information 4th Annual Internet Conf. "Of the People, By the People, For the People" Electronic Access to Government Information The Fawcett Center for Tomorrow Columbus, Ohio April 3, 1995 This conference is open to anyone interested in government information and resources available through the Internet. PROGRAM: Welcome, Michael P. Butler, OHIONET Executive Director Construction Work on the Information Superhighway: The Development of the National Information Infrastructure, Steve Kerchoff, FEDLINK Network Librarian GPO Access - First Steps, George Barnum, Case Western Reserve University Questions and Answers for A.M. Presenters, Steve Kerchoff and George Barnum Gopher Dreamin': How to Find Government Documents on the Internet, Melanie Putnam, Reference/Government Documents Librarian, The Ohio State University GIS in an Undergraduate Environment, Jerome Conley, Map/Geology Librarian, Brill Science Library, Miami University A Fist Full of Documents, Adding a Collection to the Online Catalog, Coleen Parmer, Coordinator Government Documents, Bowling Green State University "Ohio Government at Your Fingertips" via HANNAH Online, Linda Briggs, Sales and Training, HANNAH Online You CAN Get There from Here: Resource Sharing and Access to Government Information, Mary Prophet, Head of Documents/Reference Librarian, Denison University Exploring the World Wide Web "Text" and "Graphical" Browsers, Gail Junion-Metz, President, Information Age Consultants Hands-on Lab FOR COMPLETE INFORMATION, CONTACT: OHIONET. Voice: (614) 486-2966 ********************************************************** IV. PROJECTS Fr: Susanne M. Humphrey Re: IR-Related Dissertation Abstracts - September 1994 Selected IR-Related Dissertation Abstracts Compiled by: Susanne M. Humphrey, National Library of Medicine, Bethesda, MD 20894 The following are citations selected by title and abstract as being of potential interest to the Information Retrieval (IR), resulting from a computer search, using the CDP/Online system, of the Dissertation Abstracts International (DAI) database produced by University Microfilms International (UMI). Included are accession number (AN); author (AU); title (TI); degree, institution, year, number of pages (IN); UMI order number (DD); reference to the published DAI (SO); abstract (AB); one or more DAI subject descriptors chosen by the author (DE); thesis adviser (AR); and dates associated with the monthly update file (UP). 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: 1-800-343-5299; fax: 313-973-1540. 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 AAI9423984 AU Meyers, Thomas B. TI PRESERVING THE ARCHAELOGICAL RECORD: AN ARCHIVES MANAGEMENT PROGRAM FOR ARCHAEOLOGICAL DOCUMENTATION. IN Thesis (PH.D.)--UNIVERSITY OF MISSOURI - COLUMBIA, 1993, 316p. DD Order Number: AAI9423984. SO Dissertation Abstracts International. Volume: 55-04, Section: A, page: 1011. AB The research value of reposited archaeological collections is diminishing rapidly through the deterioration and loss of the associated collection documentation. Inadequate records management controls and a lack of knowledge of archival preservation techniques on the part of collection managers are the primary causal factors for this loss. Recent curation needs assessment studies conducted for the U.S. Army Corps of Engineers, the U.S. Department of Defense, and the U.S. Fish and Wildlife Service revealed the nature and extent of this problem and provided the foundation for developing and implementing a program for the management of this information. A collection management program for archaeological documentation involves three phases. The first phase consists of identifying the various archaeological projects which have been conducted under the auspices of a given agency and then locating project reports for work performed. The lack of comprehensive report libraries and/or databases, the scarcity of report copies, and the various formats in which reports are issued make the identification and recovery of this information time consuming and difficult. The identification, location, and recovery of all original documentation associate with these projects, however, cannot be achieved until this step has been completed. Phase two involves the application of archival preservation techniques which will ensure the long-term survival of these records. Information on these techniques, however, is not readily accessible to those responsible for the preservation of archaeological collections. Additionally, each documentation format has its own unique preservation requirements and these requirements must be adhered to if irreversible document damage is to be avoided. An efficient archives management plan should not only preserve the collection documentation, but also facilitate its usage. The last phase of the management plan, therefore, consists of the arrangement and description of the assembled documentation collections. The need for continued association of the documents with the recovered artifacts, however, necessitates modification of traditional archival management systems. A newly developed system of arrangement and description accommodates this requirement. DE Anthropology, Archaeology. Library Science. AR O'Brien, Michael J. UP 9409. Revised: 940930. AN AAI9424127 AU Lung, Chung-Horng. TI AN ANALOGY-BASED DOMAIN ANALYSIS METHODOLOGY (SOFTWARE RESCUE). IN Thesis (PH.D.)--ARIZONA STATE UNIVERSITY, 1994, 252p. DD Order Number: AAI9424127. SO Dissertation Abstracts International. Volume: 55-04, Section: B, page: 1519. AB Software reuse has the potential to increase productivity, improve quality, and reduce risk. Reusability is regarded as a main research subject in the area of software engineering. Domain analysis is a prerequisite for effective software reuse. Domain analysis is an expansion of the conventional requirements modeling. Domain analysis is defined as the process of identifying relevant information of a class of similar applications in a particular domain rather than just a specific system. However, domain analysis is suited to well-understood domains and is limited to a particular application area. Analogical approaches for software reuse, on the other hand, often occurs between different domains. Analogical problem solving is a process of transferring knowledge from a well-understood domain to a new problem area. Analogy can facilitate software reuse for poorly-understood problems or new application areas. But current research on software analogy overlooks the importance of domain analysis for the base domain and does not consider some critical aspects of software dynamics and analogy concepts reported in the literature. This dissertation presents an approach to software reuse, which integrates domain analysis and analogy methods. The approach is called analogy-based domain analysis (ABDA) methodology. In ABDA, domain analysis and software analogy play complementary roles. Domain analysis is regarded as a process to identify and supply necessary information for analogical mapping and transfer. Software analogy can provide the domain analyst with similar problems to start the domain analysis process. The dissertation emphasizes domain modeling techniques. A classification scheme for domain models is also presented. The methodology is then illustrated with case studies. Evaluation and classification of the methodology is conducted based on sets of criteria developed for requirements modeling, analogy, and domain analysis. DE Computer Science. UP 9409. Revised: 940930. AN AAI9424404 AU Shultz, Charles Richard, Jr. TI PRODUCTIVITY CONSIDERATIONS FOR ONLINE HELP SYSTEMS (TASK COMPLEXITY, SEARCH MECHANISM). IN Thesis (PH.D.)--UNIVERSITY OF NORTH TEXAS, 1994, 203p. DD Order Number: AAI9424404. SO Dissertation Abstracts International. Volume: 55-04, Section: B, page: 1522. AB The purpose of this study was to determine if task type, task complexity, and search mechanism would have a significant affect on task performance. The problem motivating this study is the potential for systems online help designers to construct systems that can improve the performance of computer users when they need help. Task type (mathematical and verbal) and task complexity (high complexity, low complexity) and search mechanism (keyword, semantic) were the independent variables. The dependent variable was task performance. The surrogate was task accomplishment measured by the score the subject received for the task set assigned. The researcher conducted a protocol analysis to determine the words and phrases users expect when searching for help information. From the results of the protocol analysis, the researcher constructed a semantic online help system. This research compared the traditional keyword based online help with the semantic system by assigning tasks to users where the users had to use the online help systems to accomplish the tasks. The researcher graded the users on performance. This study compared the two performance levels to identify any significant differences. The results of the research support the hypothesis that the type of online help search mechanism made a significant difference in the performance of the subjects. Task complexity was not a significant factor in affecting performance. Task type was also a significant variable in performance. Researchers will benefit by using the framework developed in this experiment. End users will benefit from the higher level of productivity provided through the use of the newer online help systems. DE Computer Science. AR Vanecek, Michael. UP 9409. Revised: 940930. ********************************************************** IRLIST Digest is distributed from the University of California, Division of Library Automation, 300 Lakeside Drive, Oakland, CA. 94612-3550. Send subscription requests and submissions to: NCGUR@UCCMVSA.UCOP.EDU Editorial Staff: Clifford Lynch calur@uccmvsa.ucop.edu Nancy Gusack ncgur@uccmvsa.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@UCOP.EDU. 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@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 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.