Representation & acquisition

Ole Jakob Mengshoel <Ole.Mengshoel@delab.sintef.no>

Mail folder: Ontolingua Mail
Next message: kamp@informatik.uni-hamburg.de: "Problems translating to loom"
Previous message: Inge Nordb|: "physical device ontology"

Date: Wed, 16 Dec 1992 06:17:08 +0000
X400-Originator: Ole.J.Mengshoel@delab.sintef.no
X400-Recipients: non-disclosure:;
X400-Mts-Identifier: [/PRMD=uninett/ADMD= /C=no/;921216131708]
X400-Content-Type: P2-1984 (2)
Content-Identifier: 1138
Conversion: Prohibited
From: Ole Jakob Mengshoel <Ole.Mengshoel@delab.sintef.no>
Message-id: <"1138*/G=Ole/S=Mengshoel/OU=delab/O=sintef/PRMD=uninett/ADMD= /C=no/"@MHS>
To: ontolingua <ontolingua@ksl.stanford.edu>
Cc: kaw@swi.psy.uva.nl
Subject: Representation & acquisition

We are interested in the representation of temporal, uncertain 
and incomplete knowledge in an interchangeable format.  The 
hypothesis is that this approach will be of help in interchanging 
knowledge among disparate knowledge acquisition and modelling 
techniques as well as among reasoning schemes.  

In this context we have found the knowledge sharing effort very 
interesting and inspiring - in particular the Ontolingua and KIF 
working groups.  There does not yet seem to be very much work 
specifically adressing the three areas mentioned above, however.  

Work that I am aware of, after a quick browsing through of some of 
the most central documents, is the following: 

- Temporal knowledge:  The Bibliography Ontology in Ontolingua. 

- Uncertain knowledge:  ? 

- Incomplete knowledge:  The :default-constraints assertions and 
  indexical knowledge in Ontolingua,  nonmonotonic inference rules in 
  KIF. 

It would be very interesting to know if there is central, existing work that 
I've missed (I suppose so), or if others are making investigations 
in these areas.  I know extensions in two of the three areas are being 
considered, cf. this quote from "The DARPA Knowledge Sharing Effort: Progress 
Report", p. 3: "Extensions under active considerations include support for 
uncertain knowledge and contexts, and additional support for default 
knowledge." 

Thanks in advance for any hints, pointers to literature or comments on the 
points mentioned above.  

Regards, 

Ole Jakob Mengshoel.