Predoctoral Trainee

Frank P. DiMaio Computer Sciences Doctoral Training Program Dept. of Computer Sciences Faculty Supervisor: Jude Shavlik Email: dimaio@cs.wisc.edu Computer Sciences web page: http://www.cs.wisc.edu/~dimaio/ 265-5693

My main research interest is in applying techniques from machine learning and computer vision to automate the interpretation of X-ray crystallographic electron density maps. Given the topology of a protein (determined from its sequence), a flexible model of protein can be built. Finding the optimal configuration of this deformable model in a density map gives the location of each atom in the protein. My work involves the construction of such a model and implementation of its matching function. Machine learning is used to determine the individual parts of the model.

I am also interested in Inductive Logic Programming (ILP) - specifically, in algorithms for dealing with extremely large problems. In problems with prohibitively large search spaces, traditional search algorithms like branch-and-bound are infeasible. In addition, empirical evidence suggests that heuristic searches like hill-climbing and A* perform poorly. Instead, the approaches that have enjoyed the most success in handling large ILP search spaces are those based upon stochastic methods. My work uses stochastic clause generation as a tool in understanding the shape of the space of possible clauses. By finding promising areas in this space, the stochastic search is guided to explore better regions in the space of possible clauses.

Resume

Resume 2002 (.doc format)

Publications

DiMaio, F. and J. Shavlik (2003). Speeding Up Relational Data Mining by Learning to Estimate Candidate Hypothesis Scores. Proceedings of the ICDM Workshop on Foundations and New Directions of Data Mining, Melbourne, FL.

DiMaio, F., J. Shavlik and G. Phillips (2003). Using Pictorial Structures to Identify Proteins in X-ray Crystallographic Electron Density Maps. Working Notes on the ICML Workshop on Machine Learning in Bioinformatics, Washington, D.C.

DiMaio, F., J. Shavlik and G. Phillips (submitted). "Interpreting Crystallographic Density Maps: A Challenging Case Study for Image Data Mining."

Search for publications by Frank P. DiMaio (Pub Med, Cite Seer)

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.