Expanding the Information and Data Management (IDM)
Research and Education Community

October 2 (8:30am) - Oct. 4 (noon), 2000
Hotel Roanoke
Roanoke, VA


See: Final report of IDM'99, including discussion of this meeting.

Links To Important Information:

Workshop Purpose

The purpose of this workshop is to bring together IDM practitioners, algorithm specialists, and tool developers to briefly summarize the state of the art in IDM and to map out a support infrastructure for the larger IDM research and educational community. Specifically, the workshop will prepare recommendations to serve the IDM community through online resources (e.g., IDM portal, digital library, Web site) that aid research, development, and education about IDM-related fields.

Related Activities

Related activities have been adopted in many diverse communities with encouraging results. For example, the Collected Algorithms, GAMS and Netlib facilities pioneered experimental investigations in the field of mathematical software. Repositories and testbeds at the community level have become accepted forums for disseminating experimental results. Software libraries and support for software testing are well developed in some research communities. Other related work includes the Protein Data Bank, GenBank, and the Quantum Chemistry Program Exchange.

IDM Issues Related to Infrastructure

The workshop will seek to explore these issues in various key areas of information and data management. Important issues include modeling the experimental process of defining a population of test problems, schema management, determining problem features most relevant to algorithm analyses, data set modeling, experiment management, and analyzing the applicability of algorithms and tools in different situations. Recommendations will be developed regarding mechanisms for building and maintaining infrastructure, including sources and amount of funding required. Discussion, focused on information and data management, will deal with issues such as: (a) test collections of audio/video; (b) software collections; (c) courseware collections; (d) linguistic/dictionary collections; (e) needs for interoperability (e.g., query translation); (f) metrics; (g) role of collaboratories; (h) technology for shared repositories; and (i) an IDM Web site (a portal, that should be continuously updated). Thus, in case (b), we will consider what types of software (experimental, demo, free, shareware, ...) should be emphasized; and what layers of services should support that software (e.g., a recommender, a match-maker, a collaboration facilitator).

Workshop Attendees and Program

Attendees: 30 or more attendees from the IDM community are expected to attend, including PIs and co-PIs on IDM-supported projects, both those who have developed algorithms/tools/data sets and those interested in using such resources (for research and/or educational purposes). If you are interested in participating, please send a position statement to Prof. Edward A. Fox at fox@vt.edu. In addition, partners will be invited from industry and other related domains. Partners from industry are encouraged to recommend ways that PIs could seek free/heavily discounted software, collaboration, summer jobs for students, etc. (as discussed at the 1998 Industry-Academia workshop).

Format: The format of the workshop will include plenary talks by key people in the field, and three breakout sessions, concentrating on:
  1. algorithms for manipulating, extracting schema from, and querying Web data (XML, DTDs, semi-structured formats).

  2. algorithms for data sets of massive dimensionality.

  3. methodologies, infrastructure (system level issues) for enabling community-level testing, evaluation and computation facilities.

These areas have been chosen for their currency and immediate relevance to the IDM specialist. For example, area 1 is increasingly gathering attention in Web site management, content personalization, and the design of internet portals. Area 2 has relevance to the large-scale information retrieval, multidimensional data mining, and knowledge discovery communities. And finally, the third working group will help address the role of superstorage systems for designing testbeds, reuse methodologies, automated experiment management, and the role of recommender systems to aid in automated algorithm selection. All together, workshop activities should lead to an operational plan for establishing and maintaining information resources that will support the large IDM research and education community.

Advance Preparation: In advance of the meeting, attendees will study the IDM program description, PIs workshop reports, IDM awards, and other relevant activities in industry and elsewhere. Based on this, attendees will prepare short reports to provide a platform for recommending the infrastructure that would enable the IDM community to be aware about relevant information.

Reporting

The recommendations of the workshop will stimulate community-wide efforts such as (i) to advance the state-of-the-art in sharing of IDM software, and (ii) to serve and extend education about IDM systems and algorithms. In addition to reporting on a WWW site (that will ultimately feed into an IDM Web site), various publication venues will be pursued. For example, with regard to aspects related to IR, there is SIGIR Forum. Also, with regard to tools that are of pedagogical value, a special issue of the new ACM Journal of Educational Resources in Computing (JERIC) will be scheduled.

Workshop Coordinators

Edward A. Fox
Department of Computer Science
Virginia Tech, VA 24061
fox@vt.edu

Layne T. Watson
Department of Computer Science
Virginia Tech, VA 24061
ltw@cs.vt.edu

Naren Ramakrishnan
Department of Computer Science
Virginia Tech, VA 24061
naren@cs.vt.edu

Steering Committee

Robert M. Akscyn, Knowledge Systems
Michael Berry, UTK
Alfonso F. Cardenas, UCLA
Don Kraft, LSU
W. Bruce Croft, U. Mass., Amherst
Edward A. Fox, Virginia Tech
Richard Furuta, Texas A & M
Alon Levy, University of Washington
Alberto Mendelzon, Univ. of Toronto
Gultekin Ozsoyoglu, CWRU
Naren Ramakrishnan, Virginia Tech
Layne T. Watson, Virginia Tech
Ellen Voorhees, NIST
Clement Yu, U. Illinois, Chicago