A cooperative coevolutionary algorithm for instance selection for instance-based learning

This paper presents a cooperative evolutionary approach for the problem of instance selection for instance based learning. The presented model takes advantage of one of the most recent paradigms in the field of evolutionary computation: cooperative coevolution. This paradigm is based on an approach similar to the philosophy of divide and conquer. In our method, the set of instances is divided into several subsets that are searched independently. A population of global solutions relates the search in the different subsets and keeps track of the best combinations obtained. The proposed model has the advantage over standard methods that it does not rely on any specific distance metric or classifier algorithm. Most standard methods are specifically designed to k-NN classifiers, however our proposal can be used with any classifier and may benefit from any specific bias of that classifier. Additionally, the fitness function of the individuals considers both storage requirements a...

• Type: Journal
• Year: 2010
• Where: ML
• Authors: Nicolás García-Pedrajas, Juan Antonio Romero del Castillo, Domingo Ortiz-Boyer
• Keywords: Algorithm, Machine Learning, ML 2010, Standard Methods, Storage Requirements