ADR-Miner: An ant-based data reduction algorithm for classification
- 1 May 2015
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in 2015 IEEE Congress on Evolutionary Computation (CEC)
- p. 515-521
- https://doi.org/10.1109/cec.2015.7256933
Abstract
Classification is a central problem in the fields of data mining and machine learning. Using a training set of labelled instances, the task is to build a model (classifier) that can be used to predict the class of new unlabelled instances. Data preparation is crucial to the data mining process, and its focus is to improve the fitness of the training data for the learning algorithms to produce more effective classifiers. Two widely applied data preparation methods are feature selection and instance selection, which fall under the umbrella of data reduction. In this paper, we introduce ADR-Miner, a novel data reduction algorithm that utilizes ant colony optimization (ACO). ADR-Miner is designed to perform instance selection to improve the predictive effectiveness of the constructed classification models. Empirical evaluations on 20 benchmark data sets with three well-known classification algorithms show that ADR-Miner improves the predictive quality of the produced classifiers. The non-parametric Wilcoxon signed-ranks test is employed to determine statistical significance.Keywords
This publication has 23 references indexed in Scilit:
- Multiple pheromone types and other extensions to the Ant-Miner classification rule discovery algorithmSwarm Intelligence, 2011
- Editorial survey: swarm intelligence for data miningMachine Learning, 2010
- Handbook of MetaheuristicsPublished by Springer Science and Business Media LLC ,2010
- Handling continuous attributes in Ant Colony Classification algorithmsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2009
- Classification With Ant Colony OptimizationIEEE Transactions on Evolutionary Computation, 2007
- Data mining with an ant colony optimization algorithmIEEE Transactions on Evolutionary Computation, 2002
- Advances in Instance Selection for Instance-Based Learning AlgorithmsData Mining and Knowledge Discovery, 2002
- Reduction Techniques for Instance-Based Learning AlgorithmsMachine Learning, 2000
- Lazy LearningPublished by Springer Science and Business Media LLC ,1997
- Instance-based learning algorithmsMachine Learning, 1991