atasha Noy, PhD ’98, was among
the first generation of computer
science researchers to study ontologies

Today, she is a senior research scientist
at the Stanford Center for Biomedical
Research and in the National Center for
Biomedical Ontologies (NCBO) at Stanford
University, where she works on Protégé,
an ontology design framework with 80,000
users worldwide. She is a thought-leader
in the development of the Semantic Web,
which will enable computers to do much
of the searching and information retrieval
that now requires human thought. “

The Web today is for human consumption,”
Noy says. “A human will probably always
need to be in the loop, but it would be nice if
computers could do more of the work for us.”

Ontologies are used to structure information
and make it searchable. They can be applied
to everything from products on a consumer
Web site to records in a medical database.
While they were once the exclusive domain
of computer science researchers, they are
now being created by people in the domains
that use them, whether that means business
people, medical professionals, or artists.

“Each person sees the world and describes
the world differently,” Noy says. “So
ontologies have very different structures
depending on who creates and uses them.”

One of Noy’s biggest claims to fame in the
world of ontologies and the Semantic Web
is her development of Ontology Develop-
ment 101, a tutorial that has become the

de facto guide for anyone who needs to
develop an ontology or understand the field.

She is also well known for her work in
ontology education through professional
conferences and workshops. “Natasha has
played an incredible leadership role in the
medical ontology community by organizing
workshops,” says Professor Carole Hafner,
Noy’s PhD advisor. “Her work has touched
many, many people.”

Among the conferences she has cochaired
is the 2007 International Semantic Web
conference, the main research conference
for the Semantic Web community. She
also co-organized a number of workshops
at major conferences on such topics as
semantic integration, collaborative knowledge
constructions, and ontology management.

when she completed her dissertation on
ontology design for experimental sciences
and use of ontologies in knowledge-based
information retrieval.

Her current work at NCBO (one of the
National Institutes of Health’s National
Centers for Biomedical Computing) will
enable users to discuss and evaluate
ontologies collaboratively.

“An ontology is an artifact that is almost as
hard to rate as a book.” Noy says. “You
can say if it is logically consistent, but
you cannot have computable means of
determining whether or not a particular
ontology is good for your task. The most
helpful information is whether the ontology
was used successfully by someone else for
a similar task.

The NCBO repository will provide a forum
for users to rate and critique various
ontologies and to provide input into what
would improve their usefulness.

“We’re working on how to facilitate, how
to reach a consensus,” she says. “Or, if
no consensus can be reached, how to
document that.”