HotAIR - COMING EVENT -- Articulate Ontologies

HotAIR LOGO

COMING EVENT--
Articulate Ontologies

Poeticality in the world of computer science

by Alice Shirell Kaswell

Despite their reputation with the general public, computer scientists can be masters of clear, concise language. Here is a good example. And you can -- and perhaps should! -- go see it.

Jargon for Laypersons, But Poetry for Their Peers

When computer scientists speak and write for laypersons, many of them use techno-jargon.

But when they talk to each other, computer scientists use simple, plain, easy-to-understand words and sentences. Typically, the clarity is stunning. It verges, many believe, on the poetical.

The "good stuff" -- the simple, clear version of computer science -- is generally hidden from the public. Perhaps this is due to some ancient, honored tradition. On occasion, though, computer scientists permit the public to read and hear the real language, the real poetry, thereal simple person-to-person, soul-to-soul communication of their basic thoughts.

Such an occasion is coming soon on, Thursday, February 8, 2001, from 4 to 6 p.m., as part of the UCLA Computer Science Department's Weekly Seminar Series Hosted by Dr. Alfonso Cardenas. It will be held in the room they call "3400 B.H."

Here is the description of the seminar. The words ring with simple beauty, and with clarity:

An Algebraic Approach to Articulate Ontologies
for Information Integration


Gio Wiederhold et al.
SKC project
Computer Science Department
Stanford University

Abstract
> > The many information resources on the web are created by autonomous
> > providers. We find these resources potentially valuable, but cannot expect
> > that the contributors will use a consistent vocabulary. While integrators
> > would benefit from consistency among sources we access, it is typically not
> > important for the sources to be consistent with the external world, and in
> > many cases, because their objectives and scopes differ among them, even
> > inefficient to be mutually consistent. In fact, the generalization of
> > mutual consistency implies global consistency, and such a state would take
> > an infinite time to achieve, and even if achieved, an infinite amount of
> > effort to maintain. The words we use are an expression of our knowledge
> > about the world, and the world and our knowledge about it evolves
> > continuously, requiring changes in terminologies, and hence maintenance of
> > their reification.
> >
> > The semantics of diverse sources are captured by their ontologies, the
> > collection of terms and their relationships as used in the domain of
> > discourse for the source. When sources are to be related we rely on their
> > ontologies to make the linkages. The traditional task of builders of
> > ontologies has been to define a single ordered structure for the domain
> > terminology, in the form of a tree or lattice. Within a limited and
> > circumscribed domain consistency can be achieved, although maintenance by
> > local domain experts will be required.
> >
> > Many applications will need information from multiple ontologies. To allow
> > shared use one has to define the articulations, namely the points where the
> > meanings of terms in distinct ontologies match, overlap, or intersect.
> > These intersections will depend on the application context. The approach in
> > our SKC project is to provide tools that define articulations that would
> > benefit an application. Now these applications will benefit from multiple
> > reliable and perhaps deep local ontologies.
> >
> > Information from articulated ontologies is acquired for use as needed. We
> > define contexts to be our unit of encapsulation for source and articulated
> > ontologies, and use a rule-based algebra to compose novel ontological
> > structures. Creating a sound algebra encompassing the required operation
> > allows manipulation and composition of the process.
> >
> > Composition of independent, specialized ontologies makes the issue of
> > semantic mismatch explicit. Our algebra has hence to support rules that
> > resolve mismatches dynamically. Once these rules exist we can reuse them,
> > and achieve maintainability of the resulting system. The required
> > articulation knowledge is seen as being held and maintained by an
> > application support specialist. Scalability becomes a dynamic concept,
> > rather than one which focuses on the growing number of rules and concept
> > frames.
> >
> > Our SKC project is still in progress. We have built some tools to aid in
> > building articulations, and can show some examples where articulated
> > composition can achieve results that would not be feasible in massive
> > systems.

For further information (directions and the like) about the seminar, get in touch with Cynthia Bautista <cynthia@cs.ucla.edu>.

We Recommend This

We heartily recommend this event. If you have friends who are non-scientists -- especially friends who mistakenly believe that scientists like to speak in unintelligible jargon and use much, much more complicated language than they have any need to -- bring those friends along. They, and you, will have a crackling good time!

(Thanks to Glenn Glazer for bringing this to our attention.)

This is a HotAIR exclusive feature. For a complete list of the others, see What's New.