Terminology and
Ontologies Track
U.S. Department of Defense
Session: Introduction to Terminology Management
8:40 - 10:00 AM, The Basic Terminology and Principles of Terminology
Management as Introduced in ISO 1087 and ISO 704
Sue Ellen Wright
Kent State University
Chair, USA ANSI TAG, TC 37
This presentation outlines the modeling variance that exists between
terminological and lexicographical resources, laying a groundwork for
terminology management. It will provide a systemic description of the
concepts in the field of terminology and clarify the use of the terms in
this field. The compilation of
vocabulary in ISO 1087 provided a forum for analyzing, discussing and
coordinating key concepts found in ISO/TC 37 standards. This standard is addressed not only to
standardizers and terminologists, but also to anyone involved in terminology
management, as well as to the users of terminologies. The principles outlined
here are valid, however, also for the definition of data element concepts and
the classification of concepts treated in ontologies and taxonomies. For
further information on this and other language industry standards, see: http://appling.kent.edu/ResourcePages/LTStandards/Chart/standards.chart.htm#Locale
Terminology management is multidisciplinary and draws support from a number of fields (e.g., logic, epistemology, philosophy of science, linguistics, information science and cognitive sciences) in its study of concepts and their representations in special languages. It combines elements from many theoretical approaches that deal with the description, ordering and transfer of knowledge. This presentation provides an introduction to ISO 704 Terminology work - Principles and Methods and is intended to standardize the essential elements for quality work in terminology management. The general purpose of the standard is to provide a common framework for documenting terminologies, i.e., the sets of terms used in special subject fields. As such, it provides guidance for establishing consistent approaches within an organization or among individuals involved in creating terminological resources. This presentation will include references to case studies involving the application of the principles articulated in ISO 704.
10:00 - 10:30
Break
10:30 -
Session: Terminological Metadata
and Interchange
10:30 -
11:15 AM, A Meta Model to Represent
Terminology Data Collections
Loria Labs, Nancy France
(Laboratoire lorrain de recherche en informatique et ses applications)
Chair, TC 37/SC 4 - Language
Resource Management
This presentation provides an introduction to ISO 16642 - Meta model for representing terminology data collections. This standard specifies a metamodel to provide guidance on the basic principles for representing the data that are stored in terminology data collections. This metamodel defines a framework for International Standards, dealing with more specific aspects of design, processing, dissemination and interchange of terminological data, e. g. ISO 12200, ISO 12620, and non-ISO standards that conform to ISO 16642, such as the LISA TBX TermBase eXchange standard. This presentation also introduces the data categories in ISO 12620 for recording terminological information in both computerized and non-computerized environments independent of software applications or hardware environments in which these data categories are used. (The new ISO 12620 will be treated in more detail later with regard to its implementation as an RDF-based metadata registry.) This presentation will also touch on work in progress in ISO TC 37/ SC 3 and SC 4.
11:15 - Noon, Terminology Data Collection Tools and Interchange
Alan K. Melby
Brigham Young University, Provo, UT
Coordinator, SALT project; chief coordinator, TBX
This
presentation provides an introduction to terminology data collection tools and
interchange. It addresses efforts to use XML to support terminology management.
It will briefly explore the revision of ISO 12620, which is
currently being developed as a metadata registry and is maintained in an RDF
resource designed for multi-level, multi-format access and manipulation of the
data categories via a cross-platform, Java-based data editor. The presentation
will proceed to address the LISA TermBase eXchange standard (TBX), which serves
as a logical companion standard to LISA's Translation Memory eXchange (TMX),
and provides a powerful, flexible solution to blind, neutralized (a)
interchange, (b) dissemination, and (c) data analysis.
The TBX standard is a
terminology markup language (TML) that is fully compliant with ISO 16642
specifications for the terminology markup framework (TMF). Finally, the
discussion will introduce the viability of combining TBX and the OLIF standard
for exchanging information from MT lexica in order to integrate terminological
and lexicographical information in interactive information management
environments, thus ensuring consistency and univocality throughout enterprises
and across media and application boundaries. This presentation will also
address issues involving work in progress on TBX and ISO 12620.
Noon - 2 PM
Lunch
2:00-
Session : Terminologies, Lexicons,
and Ontologies
Chair TC 37/SC 2 - Terminography and Lexicography
This presentation discusses the design, preparation and use of terminological ontologies that are providing the basis for semantic interoperability among heterogeneous information systems. The desired terminological interoperability can only be achieved when the metadata level and the object data level are covered in an interactive way. In order to demonstrate the methodological issues involved, the interim results of a European R&D project in field of digital culture are presented: in the process of preparing a prototype of an interactive and collaborative platform for preraring and using cultural content, an ontology is being created (from diverse existing terminologies) that is starting from the CIDOC Conceptual Reference Model. Conclusions are drawn for applying ISO TC 37 standards in this project and for improving methods of reaching terminological interoperability.
Takeya ISOBE
CTI Engineering Co., Ltd.
Even in field construction area,
IT has been introduced in various ways by using GPS or
GIS, etc. ISO/TC127/WG2 is developing standards for
data exchange between different
machinery or between site
management systems and machinery. This presentation will
introduce the standardization
activities in TC127/WG2, especially focusing on metadata
standardization. The presentation
will cover the following topics:
1. General introduction of
ISO/TC127
2. General introduction of
ISO/TC127/WG2
3. IT in field construction
4. Advanced examples in
earth-moving machinery
5. Necessity of standardization
6. System architecture
7. Future tasks of TC127/WG2
8. Liaison with JTC1/SC32
Break
U. S. Department of Defense
This will cover the use of
ontologies for formal specification of registered data models. Parts of both may be registered in 11179
metadata registries, where they can be related to new and legacy data
elements. Discussion will include change
management and management of mappings between ontologies and registered models
and elements.
Bill Anderson
(cancelled)
Ontology Works
This presentation describes
ontology-based information engineering tools. These tools are used to engineer,
and create ontologies reflecting knowledge in a business domain. The tools can
automatically generate databases and application software components directly
from this business knowledge. Issues include:
·
Database
engineering tools using ontologies,
·
Generation
of database schema from ontologies,
·
Registration
of generated evolving schema,
·
Change
management: mapping to legacy and new elements and models
DAY 4 -
Link to Track Descriptions
Terminology and
Ontologies Track
8:30 - 9:15 AM, Natural Language processing for metadata definitions and
descriptions
Eduard Hovy
Information Sciences Institute
University of Southern California
This presentation discusses
linguistic considerations and natural language processing in the development
and comparison of ontologies and related metadata elements, focusing on
definitions and descriptions.
Experiences with efforts to semi-automatically map across ontologies,
extract ontological information from online text and dictionaries, and the uses
of applications such as question answering or text summarization to help
evaluate the quality of the result will be described.
9:15 - 10:00 AM, Overall Perspectives on Current Ontology Work
Leo Obrst
MITRE
This presentation will describe
current approaches to developing ontologies and potential areas of application.
It will focus on definitions, problems, issues with respect to ontologies.
10:00 - 10:30 AM
Break
10:30 -
Mala Mehrotra
Pragati Synergetic Research Inc.
Metadata
specification requires the declaration of an underlying ontology. This
presentation will address development of technologies for building and
harmonizing axiomatized ontologies in DARPA applications. In particular, we
will present work from DARPA's Rapid Knowledge Formation project where the
focus is on building high quality large KBs, quickly and reliably. The
underlying ontology plays a key role in the quality of the axioms formulated
because of the over-generalization or over-specialization of ontological
concepts. We will also present work from an ONR project which involves building
reusable ontologies for multi-agent command and control systems for the Navy
that can address various problem-solving aspects of command and control
systems. These systems need to reconfigure themselves in real-time aspects
according to various problem solving aspects, such as, logistical versus
tactical, perspectives.
The Multi-ViewPoint
Clustering Analysis (MVP-CA) tool, built by Pragati, Inc., aids in checking
conceptual aspects of the knowledge-based software either during development or
post development stages by discovering concept islands in the KBs through
clustering techniques. The clusters helps one study the design of the ontology,
from the standpoint of the how the ontology has been used, as opposed to how it
has been declared. We believe that the basis for formulation and placement of
concepts in the ontology hierarchy should lie in understanding the context
in which the concepts will get used. We believe our analysis can contribute
significantly in specifying some unique and basic aspects of data element
composition and classification that is being addressed by various parts of the
ISO/IEC 11179 standards.
Karen Eliason
Microsoft
Taxonomies and ontologies to support searching,
browsing and schema sharing.
Noon - 2 PM
Lunch
Peter Wittenburg
Max-Planck-Institute for Psycholinguistics
This presentation covers
the use of lexicons. It describes an important metadata initiative for language
resources, in the framework of European projects. This will first outline the
needs of interoperability based on a modular approach arguing from concrete
applications. Second, it will explain a new type of controlled vocabulary
service, argue for central metadata registries and the need of describing all
aspects of metadata sets to make them available to scholars of other
disciplines for re-usage and building enhanced services.
Thomas Bandholtz
SchlumbergerSema
Use case report on a R&D project
implementing a Topic Map as a Web Service (Research project
UFOPLAN-Ref. No. 20111612, promoted by BMU/UBA,
of the Federal Agency of Environmental Protection in
Information Network), but the service is provided for general usage in the
scope of environmental
information in the "Semantic Web". The system delivers multilingual
taxonomy services, including autoclassification
features, based on a Topic Map (ISO 13250). The Topic Map contains and
integrates a
thesaurus, a gazetteer, and a chronology (approx. 100.000 terms).
Break
Harry Delugach
Univ. of Alabama in Huntsville
As researchers consider differing
proposals for representing and manipulating ontologies, there are a number of
issues which concern practitioners who wish to build practical systems to use
the ontologies. There are particular features that one can expect to be found
in a usable system, regardless of whether they are made part of a current
standard or not. This presentation will outline some of the considerations that
standards committees ought to bear in mind as they make their decisions. Some attention
will be paid to what is needed for the semantics of a representation,
especially so that it will support automated manipulation and knowledge-based
inferencing. Some examples will be presented that will illustrate different
aspects of what a practitioner would want from a standard in order to solve
problemsFormal knowledge representation for metadata registries
John Sowa
Concept Technology
This presentation describes efforts
to construct an upper level ontology. It addresses the current approach and
controversies that arise around the notion of creating an upper level ontology.
DAY 5 -
Link to Track Descriptions
Terminology and
Ontologies Track
This presentation describes the Common Logic framework--using KIF as an
example--and relationships between KIF and Conceptual Graphs (CGs), Resource
Description Framework, and XML.
Break
Pat Hayes
This presentation covers issues relating to the w3C Semantic Web initiative, including the logical foundations of RDF, other logic-like notations and their relationship to Common Logic.
End of Open Forum 2003