Re: WEBONT "HOMEWORK" (DUE DATE approaching!)

WHY: The Problem

Context: We are working on semantic annotations of images of art
objects. The purpose is to support both indexing and search through
ontologies.  There are many knowledge sources for art. We focus here
on two of these:

1. The VRA 3.0 standard for image descriptions [1], which is basically
   a refinement of Dublin Core for art-image annotation

2. The Art and Architecture Thesaurus (AAT) [2] constructed by the Getty
   Foundation, which provides a highly structured hierarchy of some
   120,000 terms to describe art objects (art categories, materials,
   styles, colors, ....) 

We want to use the WebOnt language (i) to represent the image
description template provided by VRA, (ii) to represent the AAT
hirarchy, and (iii) to link every data element ("slot") of the VRA
template to those subtrees of the AAT hierarchy where the "fillers" of
the data element can be found. For example, we want to link the VRA
data element "style/period" to the AAT subtrees representing styles
and periods.

In addition, we want to express in the ontology additional
knowledge. For example, if an indexer selects the value "Late
Georgian" for the style/period of (say) an antique chest of drawers,
we want to be able to infer that the data element "date.created"
should have a value between 1760 and 1811 A.D. and that the "culture"
is British. Availability of this type of background knowledge
increases significantly the support that can be given for indexing and
for search. 


DETAILS: What do we want to express in the WebOnt language?

[These comments are based on experiences we gained in achieving the
goals stated above with the help of RDFS, see also [3]]

1. Abstract vs. concrete classes 

EXAMPLE: In our ontology we want to define which AAT terms can act as
values
for the data element "color". AAT has an elaborate hierarchy for
colors, which is structured more or less like this (indentation =
subclass):

<color>
  <chromatic color>
    pink
      vivid pink 
      strong pink
      ....
      <intermediate pink>
        purplish pink
	  brilliant purplish pink
	  ....
	yellowish pink
	  ....
	brownish pink
    (etc.)
  <neutral color>
    white
    gray
      light gray
      ....
    black

The terms of type "<label>" are what AAT calls 'guide terms'. Their
purpose is to provide structure to the hierarchy. When we specify a
value restriction for the slot "color" of an image description
template we ideally just want to say that any subclass of the <color>
hierarchy can be used as slot filler, but we probably want to exclude
the guide terms from the value set. The difference between the guide
terms and the actual color values is close to what is being called
abstract vs. concrete classes in UML (abstract classes cannot be
instantiated, concrete classes can). Such a notion is however absent
in RDFS and in DAML+OIL.

BTW: An assumption of our work is that in order to be successful we
need to build on the (semi-)ontologies already available (like AAT),
and that we will have to map these onto a representation in a WebOnt
language. It is unrealistic to assume we can redo large-scale efforts
like AAT.

2. Aggregation

When we want to index an object such as an antique chest of drawers,
there is almost always a need to represent the part-of structure of
the object. For example, we may want to assign a style value to the
feet of a chest, e.g. "bun feet". In our view the WebOnt working group
should seriously consider introducing some (limited form) of
aggregation into the WebOnt language. If one just represents this as
another slot/relation, one loses too much of the semantics. This is
likely to be a requirement from UML people as well (aggregation has a
prominent place in UML class models).

 
3. Inter-slot constraints

In our project we want to add domain knowledge to AAT terms.  This
typically takes the form of inter-slot constraints such as:

style/period = "Late Georgian"
=>
culture = "British"
and date.created 

[Style/period, culture and date.created are all VRA data elements
defined as slots for our art-object description template.]

We could not define this constraint in RDFS and (a little to our
surprise) we saw no way of expressing it in DAML+OIL either (we could
have misread the spec, we would be glad to be proven wrong).

This type of semantical information is essential to show added value
of semantic annotations, as it makes many types of smart search
possible.

The example constraint above is of a "definitional" nature. In
prctice, we also found many similar constraints, but of a default
nature. For example, a Late Georgian chest of drawers is typically
made of mahogany wood. But default knowledge is probably a bridge too
far for WebOnt ......

Summarizing, I have placed three items on our WebOnt requirements
list:

1. distinction between abstract and concrete classes
2. some notion of aggregation
3. inter-slot constraints

Gr. Guus

[Sorry for typo's, it was written down in a bit of a rush]

[1] Visual Resources Association~Standards Committee.
    VRA core categories, version 3.0.
    Technical report, Visual Resources Association, July 2000.
    http://www.gsd.harvard.edu/~staffaw3/vra/vracore3.htm.

[2] The Art and Architecture Thesaurus
    http://shiva.pub.getty.edu.

[3] http://www.swi.psy.uva.nl/usr/Schreiber/papers/Wielinga01a.pdf

-- 
A. Th. Schreiber, SWI, University of Amsterdam, Roetersstraat 15
NL-1018 WB Amsterdam, The Netherlands, Tel: +31 20 525 6793 
Fax: +31 20 525 6896; E-mail: schreiber@swi.psy.uva.nl
WWW: http://www.swi.psy.uva.nl/usr/Schreiber/home.html

Received on Wednesday, 28 November 2001 18:22:14 UTC