|
|
KDD-2000 Sixth ACM SIGKDD International Conference on Informed Knowledge Discovery: Using Prior
Knowledge in Discovery Programs Bruce Buchanan University Professor of Computer Science and
Professor of Philosophy, Medicine, and Intelligent Systems University of Pittsburgh Abstract: Informed
knowledge discovery uses background information about a domain to guide a
discovery program toward finding interesting and novel relationships in a database. Background knowledge may be of several
forms including relationships already found, semantic categories, causal
preconditions, and taxonomic relationships.
Recent work on discovery in science will illustrate these concepts but
we will also argue for the domain-independence of the heuristics used. Biography: Bruce Buchanan has worked in artificial intelligence
since joining the Dendral project at Stanford in 1966. He was one of the
principals in the development of the Dendral, Meta-Dendral, Mycin, and
Protean programs at Stanford, and has continued working on symbolic learning
and data mining since joining the faculty of the University of Pittsburgh in
1988, where he is now University Professor of Computer Science and Professor
of Philosophy, Medicine, and Intelligent Systems. He is a member of the National Academy of Science Institute of
Medicine and is currently President of the AAAI. |
|
