Predoctoral Trainee

Jesse J. Davis Computer Sciences Doctoral Training Program Faculty Supervisor: David Page Email: jdavis@cs.wisc.edu 262-6601

My research is on the application of Statistical Relational Learning (SRL) to reading mammograms, to improve upon an existing Bayesian network for mammography. We augment SRL algorithms by adding the ability to learn new fields, intensionally deined in terms of existing fiels and intensional background knowledge. In database terminology, these new fieds constitute a learned view of the databse. We use Inductive Logic Programming (ILP) to learn rules which intentionally define the new fields. In a new approach, we propose an alternative method for generating views, based on the idea of constructing the classifier as we learn the rules. Rules are scored by how much they improve the classifieer, providing a tight coupling between rule generation and rule usage. The final framework extends this alternative method to leverage an initial set of features. This yields a tighter coupling between view generation and view usage.

CV

CV (.pdf format)

Publications

Search for other publications by Jesse Davis(Pub Med)

Feedback, questions or accessibility issues.
Graduate Program  |  Postdoctorate Program  |  Short Term Research Program

File last modified: August 13, 2003
Copyright © 2003 The Board of Regents of the University of Wisconsin System.