Petra Philips, Ph.D./The Australian National University

My research interests include both theoretical aspects and applications of machine learning and statistical techniques to structured natural data.

Currently, as a postdoc in Gunnar Rätsch's computational biology group, my investigations are concerned with the analysis of DNA sequences. In the projects I am working on, I do explore kernel-based classifiers for the detection of gene as well as intron boundaries. On a broader front, I am inquiring the following question: Is it possible, based on modern statistical learning methods, to build machines which out of millions of observations can grasp so much meaning as to help answering significant biological questions?

During my PhD at the Australian National University in Canberra, Australia, I analyzed the generalization ability of statistical learning machines. Before being a doctoral student, I was affiliated with the Philips Research Laboratories (Aachen, Germany), doing research on speech understanding for conversational systems, with the Advanced Telecommunications Research Laboratories (Kyoto, Japan) working on speaker adaptation, and with the Intelligent Data Analysis Group, Fraunhofer Gesellschaft (Berlin, Germany) investigating methods for the separation of linearly superposed and statistically independent signals. I received my Masters in Computer Science from the University of Karlsruhe.

Publications

Statistische Methoden (HFWU Nürtingen, SS08)