Manuela Zanda

School of Computer Science
University of Manchester
Kilburn Building
Oxford Road
Manchester
M13 9PL
UK



About Me

NEWS:
I am now a Postoctoral Fellow in Matthew Hurles' Genomic Mutation and Genetic Disease Group at the Wellcome Trust Sanger Institute, Cambridge, UK, where I am involved in a study conducted by the Type I Diabetes Genetics Consortium to investigate the role of rare and common Copy Number Variants on disease susceptibility.

You can contact me now at mz5 at sanger.ac.uk

In Novemeber 2010 I completed my PhD in Computer Science under the supervision of Dr. Gavin Brown in the Machine Learning and Optimization Group at the School of Computer Science, University of Manchester.

My PhD dissertation "A Probabilistic Perspective on Ensemble Diversity" was concerned with the ensemble learning concept of classifier diversity in both neural networks and augmented Naïve Bayes classifiers. It also presented empirical evidence that interaction information can be used to monitor diversity in classifier ensembles and that diversity occurs at different levels of interactions between base classifiers. The main research finding of my thesis was that interaction information can be used to build ensembles of averaged augmented Naïve Bayes classifiers in a more computationally efficient manner than accuracy based ensemble methods.

In December 2006 I graduated as a MSc student at the Department of Electrical and Electronic Engineering, University of Cagliari, with the dissertation "An Experimental Analysis of AdaBoost", under the supervision of Dr. Giorgio Fumera and Professor Fabio Roli. My MSc investigated to what extent the classification performance of the ensemble learning method AdaBoost can be interpreted in terms of the Tumer and Ghosh framework,that is in terms of the bias, variance and covariance decomposition of the training error.

(More) About Me

Publications

Software


Last update: 06/02/2011