I feel that the acrnyonm "ONTAC" should have some relationship to what this
group says that it is doing. (01)
I note also that there is a difference between what is discussed in the
forum and the decision that are made behind the scenes. Right? (02)
>From the Nov 2005 work plan (it would be nice if the acrnynom were spelled
out) (03)
NCOR is the academic reserach institute that Barry Smith is associated with.
I wish I knew what KOS happens to stand for.
UMLS stands for Unified Medical Language System
etc (04)
Nov. 3, 2005 (05)
ONTACWG work plan: coordinating efforts on Knowledge Classifications Systems
and interoperability (06)
This ONTACWG work plan is an outcome of discussions among ONTACWG members'
participating in the NCOR inauguration event. There will be close
coordination between ONTACWG and NCOR as part of a public-private-academic
partnership. Ideas for additions to and modifications of this work plan
should be posted to the WIKI page and/or the list (07)
Registry of ontologies and other KOS (08)
ONTACWG will collaborate on the NCOR’s planned ontology registry by
contributing a list of metadata elements through the steps listed below.
ONTACWG’s focus is on registering use cases and ontologies and other KOS
from the Federal government. This registry effort will not preclude
registering the same KOSs with other registries. (09)
The registry should be Web-based system for self-registration using a
structured data input form, possibly amended later by additional data from
the registry editors.
Step1. Collect data on desired metadata elements and on existing registries
(by email in the subgroup)
Step2. Make a table comparing the registries (DS)
Step3. Group meeting on the list of metadata elements to be recommended to
NCOR (010)
Near-Term Substantive work to be done by group members working
collaboratively (011)
Priority 1: UMLS semantic network. The first expected project is to revise
the UMLS semantic network (SN) starting with Barry Smith’s et al. definition
of basic relationship types and mapping to one or more upper ontologies.
Olivier Bodenreider and Lowell Vizenor at NLM have already begun
reorganization and mapping of the UMLS-SN to an upper ontology based on
DOLCE and BFO. Efforts by ONTACWG members to map the UMLS-SN to other upper
ontologies would be most welcome (please communicate interest to Pat
Cassidy); results would be communicated to NLM for their consideration. (012)
Priority 2: Analyze the FEA DRM in a similar manner. This still requires a
volunteer, and will be helped by a student from UMD doing an independent
study. This will be coordinated with the submission of the DRM ontology to
NCOR for testing. (013)
Priority 3: In a similar manner, create mappings of the DoD Core Taxonomy
to one or more upper ontologies. At this point, no volunteer has come
forward to take this on immediately. This part of the overall mapping
project may be delayed until after progress has been made on the first two
priorities. (014)
Priority 4: Examine the higher levels of the domain ontologies to determine
overlap based on 1 – 3 above. Comparisons of the mappings of the UMLS-SN or
other domain ontologies to multiple upper ontologies are expected to provide
information that will help determine which upper ontology or combination of
upper ontologies would be the optimal candidate for the core of the Common
Semantic Model (COSMO). Identifying similar concepts in different domain
ontologies will allow comparisons to help determine the effect of context on
details of related concepts, by examining the detailed conceptual
specifications made available by the mappings to the upper ontologies. (See
procedure at the end of this document.) (015)
Concurrent and continuing effort
Concurrent with all of the above projects: members of the full ONTACWG will
be encouraged to monitor and comment on ongoing formalization efforts to
ensure that any concepts related to their own domains of interest, when they
appear in any of the formalizations, are represented in such a manner as to
include the essential conceptual elements that make up the meanings of those
concepts as they are viewed by any interested community or individual.
Where important concepts appear to be missing or not represented as
understood by any community, recommendations for additions or changes should
be made. (016)
ONTACWG members are also encouraged to compare types of entities and
relationships between the different domain ontologies (UMLS-SN and FEA-DRM)
as the work goes on, to discover commonalities across different domains that
may be easier to identify from different community perspectives and may
escape those focusing on single domains. (017)
General areas suggested for ontology work ONTACWG members or others might
undertake (including work under Priorities 1 – 4)
Analyzing domain ontologies with respect to one of the upper level
ontologies
Exploring methods to construct mappings for interoperability
Sharing ideas for exploratory or substantive work that could lead to funded
work by ONTAWG members or others. (018)
Information sharing (019)
Resource section of the WIKI page All members are encouraged to add useful
resources with an indication how the resource can be used in furtherance of
ONTAWG' work. (020)
Information sharing on other independent formalizations of medical and other
ontologies. Anyone participating in or in contact with people participating
in such projects is requested to contact the ONTACWG list or a member of the
COSMO-WG to arrange for regular exchange of information that will avoid
duplication and allow each group to take advantage of the other's efforts
while the projects are still in progress. (021)
Previously suggested general approach for comparing upper-level ontologies
(D. Soergel) (022)
The general approach suggested earlier by Dagobert Soergel for mapping
upper ontologies is reiterated here for any further comment. (023)
1 Collect suggested upper ontologies (024)
2 Compare and determine differences
2.1 In elements (presence / absence and, more difficult, definition)
2.2 in relationships (025)
3 Try to resolve differences, creating a superstructure that incorporates
the non-contradictory parts of various schemes
3.1 By adding elements
3.2 By adding relationships (026)
4 Articulate the remaining differences so that they are clearly understood (027)
ONTACWG members are encouraged to recommend additions or modifications to
this process for mapping the different upper ontologies into a neutral
superset that was described briefly in an email from DS and is repeated
below. This could then become the basis of a proposal, that may generate
direct funding, possible conducted as part of the NCOR, with ongoing results
communicated to ONTACWG for comments. Such a project might have a project
advisory board drawn from ONTACWG members. (028)
_________________________________________________________________
Message Archives: http://colab.cim3.net/forum/ontac-forum/
To Post: mailto:ontac-forum@xxxxxxxxxxxxxx
Subscribe/Unsubscribe/Config:
http://colab.cim3.net/mailman/listinfo/ontac-forum/
Shared Files: http://colab.cim3.net/file/work/SICoP/ontac/
Community Wiki:
http://colab.cim3.net/cgi-bin/wiki.pl?SICoP/OntologyTaxonomyCoordinatingWG (029)
|