event
Thursday 07 Nov 2013: Recent Improvements to Ant Colony Classification Algorithms
Dr. Fernando Otero - School of Computing at the University of Kent
Harrison 101 15:00-16:00
In this presentation I will talk about recent improvements to Ant
Colony Optimization (ACO) algorithms for discovering comprehensible
classification models. First, I will discuss the commonly-used search
strategy used in ACO algorithms to create classification rules. Then,
I will present an improved search strategy that allows the algorithm
to search for the best list classification of rules. I will also
discuss how we can further improve the interpretability of the
classification rules discovered by ACO algorithms. Finally, I will
discuss potential research directions.
Bio: Fernando Otero is a Lecturer in the School of Computing at the
University of Kent, UK. His current research interests are
biologically-inspired algorithms, bioinformatics, and data mining. He
is particularly interested in applications of Ant Colony Optimization
algorithms to data mining problems.
