Archive Site

CLIMB Project Page

Original CLiMB home page


Tools & Prototypes

Publications & Presentations


Grant Proposal (pdf)
Grant Announcement (pdf)

Internal Documentation 


The goal of the project Computational Linguistics for Metadata Building (CLiMB) is to explore and develop computer-assisted strategies for extracting item-level metadata -- i.e., cataloging suitable for use in large-scale digital library collection projects -- from the body of existing scholarly monographic literature relating to those collections. This effort is an attempt to address the problem presented by the often prohibitive expense of creating detailed subject cataloging for items in the collections being digitized. This is especially true for the many unique image-based research collections that would be of significant scholarly and instructional interest if they were made available as part of the evolving national and international digital library.

The techniques to be developed in the CLiMB project thus offer the promise not only of improving the creation of descriptive metadata, but also of increasing access. Such metadata will be extracted from text which is in some way associated with an image, either explicitly or by topic. We will collect this metadata to explore its use for image collections. We propose a thorough and ongoing assessment of the metadata and an evaluation of its use within existing platforms.

The Columbia University project team is composed of an interdisciplinary group of librarians and computational linguists as well as research-oriented 'use experts' such as art curators, reference staff and selected faculty.

The Principal Investigator is Dr. Judith L. Klavans

Director, Center for Research on Information Access
Department of Information Services
535 West 114th Street,
New York, NY 10027
212-854-7443 (phone)
212-854-9099 (fax)
  Research Scientist
Department of Computer Science
500 West 120th Street,
New York, NY 10027
212-939-7119 (work- CS)
212-939-7110 (assistant - CS)