baseball-ont
is an ontology for describing
baseball teams, players, games, and plays.
It was developed as demonstration and training material
in an application domain generally understood by U.S. audiences.
See
here
for an example.
Approach
The basic approach is to have an ordered sequence of
events extending from the team, to the game, to innings,
to individual pitches, swings, and plays.
No aggregation (league standings, box scores, statistics, etc.)
is done in this ontology;
it's expected that properties from a baseball aggregation ontology
could be added to the
classes defined in this ontology.
Production
This ontology was developed using
OilEd 2.2a.
baseball.rdfs
is the actual OIL-in-RDFS source;
baseball-ont.daml
was exported from OilEd.
Some instances have been created using
Ontomat.
A UML representation of this ontology,
produced using
DUET,
is available
here.
Related Work
After developing this ontology,
we became aware of existing tools such as
TurboStats ScoreKeeper
(which fans can
beam
into their Palm devices
from kiosks at Pacific Bell stadium)
and
TurboStats.
Relevant material may also be available from the
Society for American Baseball Research (SABR).
Author
Mike Dean
$Id: index.xml,v 1.3 2002/04/11 14:46:56 kmbarber Exp $
