Search   |   Back Issues   |   Author Index   |   Title Index   |   Contents

Conference Report


D-Lib Magazine
November/December 2008

Volume 14 Number 11/12

ISSN 1082-9873

Networked Knowledge Organization Systems/Services (NKOS)

ECDL 2008 Workshop Report


Marianne Lykke Nielsen
Royal School of Library and Information Science, Denmark

Red Line


The 7th European Networked Knowledge Organization Systems/Services (NKOS) workshop, organized by Marianne Lykke Nielsen in collaboration with Doug Tudhope and Traugott Koch as co-organizers, took place on September 19, 2008, in Århus, Denmark as part of the European Conference on Digital Libraries (ECDL) 2008. The full day workshop was attended by 24 people.

Knowledge Organization Systems/Services (KOS), such as classification systems, thesauri, taxonomies, and ontologies attempt to model the underlying semantic structure of a domain. They serve a variety of functions: tool for representation and indexing of information and documents, knowledge-based support for information searchers, semantic road maps to domains and disciplines, communication tool by providing conceptual framework, and conceptual basis for knowledge based systems, e.g., automated classification systems. Modern digital information systems afford more options for mapping and presenting alternative orders of information than traditional physical libraries. Thus, the challenge is as much intellectual as technical when we want to develop and map knowledge organization systems that are useful and meaningful for end-users operating in complex, interdisciplinary knowledge domains. The NKOS workshop addressed some of the challenges involved in leveraging the full potential of KOS for advanced digital library applications.

The first morning session concerned the use of social tagging to improve online access to art collections. Jennifer Trant and David Bearman, Archives & Museum Informatics, Canada presented results from a large research project investigating how social tagging can contribute to the organization and retrieval of museum objects. Based on results from a study in the steve museum project,1 they concluded that tagging does contribute. Tags assigned by users are different from the established public vocabulary used by art professionals, and cannot be mined from other sources. 85% of the user tags are not found in the museum documentation, and 60% do not match established art vocabularies. Marianne Lykke Nielsen, Royal School of Library and Information Science, Denmark followed up with a general overview of trends and findings in social tagging research. The aim of the presentation was to put forward a plenum discussion of social tagging and its relation to KOS.

The topic of the second morning session was networked KOS in general. Marjorie M.K. Hlava, Access Innovations / Data Harmony, USA reported findings from a study at the U.S. Weather Channel investigating 'return on investment' (ROI) using automatic term suggestions in retrieval. The term suggestion feature expanded the original search query by a large set of synonyms providing a ROI of 10% reduction in search time. Magnus Pfeffer, University Library, University of Mannheim, Germany presented a thesaurus-based system for automatic indexing. A study of the indexing quality showed that synonyms are essential for term disambiguation and to avoid assignment of overly broad concepts.

In the session of short communications Felix Boteram and Jessica Hubrich, Fachhochschule K´┐Żln, Germany proposed a model for a comprehensive international knowledge organization system offering interoperability between KOS and meeting challenges such as language, structural and typological differences. Christoph Wieser, Salzburg Research, Germany presented the damped clustering concept method that uses a thesaurus model to assign index terms automatically to newspaper articles. The method favours the occurrence of narrower terms in the document text. Marcia Zeng, School of Library and Information Science, Kent State University, USA reported on the US NKOS/CENDI workshop, held in Washington, D.C., September 11, 2008.

The first afternoon presentations concerned terminological registries. Doug Tudhope, Glamorgan University, UK and Kora Golub, UKOLN, UK reported the TRSS Terminology Registry Scoping study that described scope and potential use of terminal registries. They analysed requirements for registry services with the overall aim of helping stakeholders understand the need for this component of shared infrastructures and to point to future research. Marcia Zeng continued by putting forward the questions: what do we need to know about terminology resources, do we need metadata descriptions at all, and is there a standardised set of metadata for terminology resources?

The last session concerned domain-specific, faceted KOS. Claudio Gnoli, University of Pavia, Italy discussed the potential of freely faceted classification for knowledge retrieval and browsing, and he showed specifically such classification's potential for systematic browsing. Marianne Lykke Nielsen presented a faceted indexing approach to marking up semantic units or components of document texts. The method seeks to improve precision in domain-specific information retrieval. Findings from a comparative study evaluating retrieval effectiveness showed improvement on average of 25% measured by nDCG (normalized discounted cumulative gain).

The NKOS presentations resulted in a wide range of comments and discussion throughout the workshop, which was an important workshop goal. Participants identified various issues and themes for continuing work:

  • Use cases and business models for shared infrastructures and registries
  • Need for evaluations and validation of KOS
  • Investigations of the role of folksonomies in retrieval
  • Need for empirical studies of end-user interaction with KOS
  • Issues and problems related to vocabulary mapping and interoperability across domain-specific KOS
  • Studies of end-user vocabulary vs. professional jargon
  • Use of structure and suggestions in social tagging

Further information on the workshop and links to the presentations can be found on the workshop website at <>. NKOS is a community of researchers, developers and practitioners seeking to enable KOS as networked interactive information services via the Internet. This was the seventh European NKOS Workshop.

In line with previous workshops authors presenting at the workshop are encouraged to submit papers for consideration in future issues of the Journal of Digital Information (JoDI) and New Review of Hypermedia and Multimedia (NRHM).


1. The project team responsible for the steve museum deliberately chose to use the lower-case "s" because "steve" 'is neither a person nor an acronym...the team wanted something simple, friendly, and easy to remember.'

Copyright © 2008 Marianne Lykke Nielsen

Top | Contents
Search | Author Index | Title Index | Back Issues
Previous Conference Report | Next Conference Report
Home | E-mail the Editor


D-Lib Magazine Access Terms and Conditions