Information Retrieval List Digest 152 (February 23, 1993) URL = http://hegel.lib.ncsu.edu/stacks/serials/irld/irld-152 IRLIST Digest ISSN 1064-6965 February 23, 1993 Volume X, Number 8 Issue 152 ********************************************************** I. NOTICES A. Meeting Announcements/Calls for Papers 1. EACL '93: Provisional Program 2. PACLING '93: Paper Topics and Registration Information C. Miscellaneous 1. UC Berkeley Library School Closing? 2. IR-L Digest Available via FTP II. QUERIES B. Requests for Information 1. Word Weights in Full Text Documents 2. Creating Bibliographic DB ********************************************************** I. NOTICES I.A.1. Fr: EACL 1993 Re: EACL93: Provisional Programme THIRD NOTIFICATION -- PROVISIONAL PROGRAMME The European Chapter of the Association for Computational Linguistics will hold its Sixth Conference in Utrecht, The Netherlands, from Wednesday to Friday, 21-23 April 1993, preceded by two days of tutorials on Monday 19 and Tuesday 20. The final programme is now being prepared, and a full time schedule will be published shortly. This announcement provides an overview of the contents of the main programme, and includes registration information and forms. Please note that the early registration period has been extended to 1 March, but note also that hotel accommodation cannot be guaranteed after this date! TUTORIALS (Monday 19 and Tuesday 20 April): Each tutorial consists of an introductory class on Monday (3 hours), and an advanced class on the same topic on Tuesday (3 hours). There are 4 tutorials: (1) Jeroen Groenendijk and Martin Stokhof (University of Amsterdam): Uses of Dynamic Logic in NL Processing (2) Hans Uszkoreit (University of Saarbruecken): Recent Developments in Unification-based NP Processing Tutorials (1) and (2) will take place in parallel. (3) Mark Liberman (University of Pennsylvania): Statistical Methods in NL Processing (4) Applications of Complexity Theory (teacher to be announced) Tutorials (3) and (4) will take place in parallel. MAIN PROGRAMME (Wednesday 21, Thursday 22 and Friday 23 April): INVITED SPEAKERS: (1) Ken Church (AT&T) (2) Johan van Benthem (University of Amsterdam) (3) Third speaker to be announced PAPERS: Below we list the topics on which papers will be presented. SYNTAX AND CL LEXICON, MORPHOLOGY DATA-ORIENTED CL AI-RELATED METHODS IN CL PARSING AND COMPLEXITY LOGIC AND CL MORPHOLOGY, PHONOLOGY, SPEECH MACHINE TRANSLATION SEMANTICS ALSO: STUDENT PAPERS POSTERS and DEMOS REGISTRATION INFORMATION NOTE THE EXTENSION OF THE EARLY REGISTRATION PERIOD!!!!!!!! Conference: The registration fee will be 275 Dutch guilders for ACL-members, and 165 guilders for students and unemployed members. Registration forms and payment should be received by 1 March 1993. The fees include one copy of the proceedings and an invitation to the reception. Addresses: General address for all communications with Programme Committee, Organizing Committee, Student Session Committee and Tutorials Coordinator: EACL93 [relevant committee/coordinator], OTS, Trans 10, NL-3512 JK Utrecht, The Netherlands. Tel: +31 30 53 63 77 Fax: +31 30 53 60 00 Email: eacl93@let.ruu.nl For information on the ACL in general, contact Don Walker (global), or Mike Rosner (for Europe): Dr. Donald E. Walker (ACL) Dr. Michael Rosner (ACL) Bellcore, MRE 2A379 IDSIA 445 South Street, Box 1910 Corso Elvezia 36 Morristown, NJ 07960-1910, USA CH-6900 Lugano, Switzerland walker@flash.bellcore.com mike@idsia.uu.ch ********** I.A.2. Fr: Dan Fass Re: PACLING-93 invited/accepted papers + registration information INVITED/ACCEPTED PAPERS AND REGISTRATION INFORMATION PACLING '93 First Pacific Association for Computational Linguistics Conference April 21-24 (Wed-Sat) 1993 The Harbour Centre, Vancouver, British Columbia, Canada CONFERENCE AIMS: PACLING '93 will be a workshop-oriented meeting whose aim is to promote friendly scientific relations among Pacific Rim countries, with emphasis on interdisciplinary scientific exchange showing openness towards good research falling outside current dominant "schools of thought," and on technological transfer within the Pacific region. GUEST SPEAKERS, TITLES AND BRIEF ABSTRACTS "An Overview of JPSG -- A Constraint-Based Grammar for Japanese" Dr. Takao Gunji, Osaka University, Japan: An overview of an ongoing project called JPSG (Japanese Phrase Structure Grammar) is presented. JPSG is an implementation of ideas from recent developments in the phrase structure grammar formalism, such as HPSG, applied to the Japanese language. Even though JPSG shares many aspects of grammatical formalization with HPSG, we have adopted a number of extensions and modifications in our development. We use an extended notion of unification -- constraint unification -- which takes into account declarative constraints, in addition to feature structures, so that the same declarative description can be used both for generation and recognition. "Industrial Strength NLP: The Challenge of Broad Coverage" Dr. George E. Heidorn, Microsoft Research, USA: To achieve apparent natural language understanding in consumer products, the underlying NLP system will have to be very robust. It will be expected to do more than a limited task in a limited domain. The NLP group at Microsoft Research is developing a system which is intended to be a central component for ubiquitous NLP. This work is still in its early stages, but we do have a system of some interest that runs on Windows 3.1 and produces reasonable logical forms for a fairly wide range of English text. This talk will describe the various facets of the work we are doing and show some of the results we have obtained to date. "Language Generation for Multimedia Explanations" Dr. Kathleen R. McKeown, Columbia University, USA: Multimedia information systems have the potential to greatly increase the effectiveness with which information is communicated. Whether language, visual media (e.g., pictures, charts, figures, etc.) or some combination are more appropriate for communication can depend on the kind of information being communicated, on user ability or background, and on the situations in which information is communicated. Our work on COMET (COordinated Multimedia Explanation Testbed) has as its goal the interactive generation of explanations that fully integrate and coordinate text and graphics, all of which is generated on the fly. In this talk, I will focus on three ways in which COMET coordinates its text and graphics: 1. cross references from text to graphics, 2. coordination of sentence and picture breaks, and 3. influence from one media on realization in another. TOPICS FOR ACCEPTED PAPERS: (Confirmation of acceptance is still to be received from the authors of some accepted papers; groupings of papers are tentative.) *** Morphology, Phonology and Prosody *** *** Parsing -- General *** *** Parsing -- Language-Specific *** *** Semantics and Cognitive Modelling *** *** Pragmatics and Discourse *** *** Natural Language Generation/Explanation *** *** Natural Language Generation/Planning *** *** Machine Translation and Machine Assisted Translation *** *** Document Structure and Language Learning Aids *** *** Information Retrieval/Extraction and Large-Scale Lexical Resources *** POSTER SESSIONS, DEMONSTRATIONS, COMPUTER FACILITIES: Approximately 6 people are being invited to present posters. Invited speakers and authors of accepted papers are being encouraged to give demonstrations of their systems. The conference is providing Mac, IBM, NeXT and SUN machines for demonstrations. Conference attendees will be provided with guest e-mail facilities during the conference. CONFERENCE REGISTRATION INFORMATION: Full registration fees for the conference, besides attendance at conference sessions and use of guest e-mail facilities, include: * copy of the conference proceedings * reception * banquet * day trip to Whistler Village, home to two of the finest skiing areas in North America, Whistler and Blackcomb Mountains; the village and surroundings are very picturesque; the village has many shops and restaurants; skiing is still good in April, weather permitting (ski passes not included in registration fee). For further information on the conference and on local arrangements, contact: Dan Fass email: fass@cs.sfu.ca PACLING '93 Publicity and Local Arrangements tel: (604) 291-3208 Centre for Systems Science fax: (604) 291-4424 Simon Fraser University Burnaby, British Columbia, Canada V5A 1S6 ********** I.C.1. Fr: Genny Engel Re: UC Berkeley Library School The School of Library and Information Studies at the University of California, Berkeley has been under academic review for some time with mixed results. Some of the review committees recommended expansion of the program as a national resource in information studies, while others recommended closure of the School in the face of the budget problems of higher education in California. Other options, such as the merger of the Information Studies program with another department on campus, have been ruled out by the review committees. The School now faces either expansion or complete closure. If the prospect of the School's closure is alarming, the prospect of its expansion is equally exhilarating. Down to a faculty of seven and an interim Dean, the School nevertheless has attracted major funding in recent years for faculty research in information systems and their future application in areas as diverse as document imaging/retrieval and the management of geographic data from satellites. While maintaining a core curriculum in reference services and in cataloging and organization of materials, the School continues to educate the information retrieval and networked information professionals and theorists of the future. The Academic Planning Board will now review the other committees' recommendations and present its findings to the Chancellor of UC Berkeley in late February. The School's Alumni Association has mounted a letter-writing campaign and welcomes other ideas for indicating support of the School. For further information, contact the School's Alumni Association or its Acting Dean: Nancy Van House, Acting Dean School of Library and Information Studies South Hall University of California Berkeley, CA 94720 Mary Jo Levy, President SLIS Alumni Association phone (415) 329-2516 fax (415) 327-7568 ********** I.C.2. Fr: Nancy Gusack Re: FTP Access to IR-L ********************************************************** II. QUERIES II.B.1. Fr: J.J. Paijmans Re: Word Weight in Full Text Documents I am currently doing research on weights of words in full-text (FT) documents (as opposed to abstracts). To do some experiments I need as many FT-documents as I can get (a few hundred to a thousand). Such a document should be marked up with LaTeX without (or with very little) home-made macros, be written in English, and have an abstract attached. Subject matter is not really important, but computer science, or IR-related documents would be nice. The documents themselves will not be published, except for some fragments of some documents to illustrate some point or other. So if you have some TeX-documents as described above, or if you know of databases of full-text documents (not necessarily in LaTeX), that I could use, please contact me by email. I have a second question: After looking in the literature on the general subject of IR of the last twenty-odd years, it strikes me that it often is not really obvious (to me) if experiments have been conducted on real full-text documents or on abstracts of the same. An example is the Salton/Zhang article in Information Processing and Management, 1986, "Enhancement of text representations using related document titles." On page 388 they refer to the "original documents" of the CACM-collection, but all examples show only abstracts. This is typical for the literature that I have seen. Am I particularly dense, or do other collegues have the same experience? Could somebody direct me to literature that explicitly tackles the differences between (weights of keywords in) abstracts and full-texts? Thanks in advance Hans Paijmans paai@kub.nl ********** II.B.2. Fr: Barbara Tyszkiewicz Re: Creating Bibliographic DB The Institute of Literary Research of the Polish Academy of Sciences intends to create and maintain a public access database containing: 1) bibliographic data concerning polish literature and its history 2) biographical data concerning writers, translators, historians of the literature, etc. 3) information about institutions of the polish literature: organizations, informal grups, periodicals, publishers, etc. The intended contents of the database shall cover currently existing bio- and bibliographical sources, in particular: Polish Bibliography of Literary (yearly from 1945), Nowy Korbut, Dictionary of Contemporary Polish Writers. At the moment we are choosing the structure of the database, software tools necessary to create it etc. We would like to know about any similar enterprises, to share the experience. Any suggestions from the prospective users are also welcome. Barbara Tyszkiewicz Institute of Literary Research, Polish Academy of Sciences ul. Nowy Swiat 72 p. 130 00-330 Warszawa, Poland. e-mail bartysz@plearn.bitnet, jurekty@melkor.mimuw.edu.pl (preferred) ********************************************************** 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.bitnet or ncgur@uccmvsa.ucop.edu Mary Engle meeur@uccmvsa.bitnet The IRLIST Archives is being set up for anonymous FTP, and access information will be provided soon! 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.