Helping computers learn to classify what they see at a glance, thanks to $275,000 in ARC funding

Published:30 October 2015

CQUniversity Professor Brijesh Verma pictured at a recent Congress on Evolutionary Computation.

Helping computers learn better to classify what they see at a glance has many real-world applications such as document analysis, robotics and medical diagnosis.

That's the goal of CQUniversity researcher Professor Brijesh Verma who has gained $275,000 over the next three years, thanks to a competitive bid to the Discovery Project program of the Australian Research Council.

The CQUni Brisbane-based academic says his project with a partner investigator (Prof Mengjie Zhang) aims to develop a novel framework for creating an optimised ensemble classifier that will improve data analysis and accuracy of many real-world applications.

"The project plans to develop and investigate novel methods for generating diverse training environment layers, base classifiers and fusion of classifiers," Prof Verma says.

It also plans to design a multi-objective evolutionary algorithm-based search to obtain the optimal number of layers, clusters and base classifiers.

"The expected outcomes of the proposed framework are advances in classifier learning.

"The final outcome will be novel methods which will bring in diversity during the learning of the base classifiers and provide an optimal ensemble classifier for real-world applications."