ERCIM News No.32 - January 1998

EU-NSF Working GroupWorking Group on Multilingual Information Access

by Judith Klavans and Peter Schauble


The first meeting of a new working group on Multilingual Information Access for Digital Libraries took place 16-18 November 1997 at the Columbia University Arden Homestead in New York under the sponsorship of the U.S. National Science Foundation (NSF) and the European Union (EU). The focus of the Multilingual Information Access working group is on the problems faced by the international community with storage, access and presentation of information in any of the world's languages.

The Multilingual Information Access working group includes researchers from information retrieval and natural language processing, two fields that are converging increasingly in the area of information access. The US participants in the meeting were Jaime G. Carbonell (Computer Science, Carnegie Mellon), Vasilieos Hatzivassiloglou (Computer Science, Columbia), Eduard Hovy (Information Sciences Institute, University of Southern California), Judith L. Klavans (Butler Library, Columbia University), Doug Oard (Library and Information Services, University of Maryland), Evelyne Tzoukermann (Bell Laboratories) and Gary Strong from the National Science Foundation. Unfortunately, Bruce Croft (University of Massachusetts) was unable to be present but sent the group a list of relevant issues for discussion. Europe was represented by David Hull (Multilingual Theory and Technology, Rank Xerox Research Centre), Carol Peters, (Institute for Information Processing of the Italian National Research Council), and Peter Schauble and Paraic Sheridan (both from ETH-Zurich, Institut für Informationssysteme).

Two primary areas of research were addressed during the first meeting: first, the problem of encoding, manipulating and displaying information in any language, and second, methods for querying, retrieving and presenting that information. The discussion concentrated on understanding user needs, identifying existing tools and resources, and prioritizing research issues for near, medium, and long-term planning. For example, there is a growing need to support access to documents in many languages with retrieval systems that can accept queries in the native or preferred language of the user. Furthermore, an accurate profile of the user's linguistic capabilities would allow retrieved information to be presented without translation when possible, but translated or summarized in another language when needed.

At the second working group meeting, scheduled for March, 1998 in Zurich, it is hoped that a representative from the Pacific Rim will also be present to enhance the group with additional perspectives. In that meeting the group is expected to complete the user needs assessment, further refine the set of research issues, and prioritize the research agenda. The long-term aim is to assist the multilingual information access community to identify research directions towards which future efforts should most usefully be concentrated. The working group plans to produce a white paper containing findings and recommendations in the summer of 1998.

Since the Multilingual Information Access working group seeks to serve the larger Digital Library community, comments and recommendations from researchers interested in these issues will be invited.

Additional information about the working group and contact information for the participants is available on the group's web site at http://www.cs.columbia.edu/~klavans/Activities/MIA/home.html or from the working group co-chairs: Judith Klavans (US) and Peter Schauble (EU).

Please contact:

Judith Klavans - Columbia University
E-mail: klavans@cs.columbia.edu

Peter Schauble - ETH, Zurich
E-mail: schauble@inf.ethz.ch


return to the contents page