Adaptive Classifier Selection System Using Context-driven Genetic Algorithm

Adaptation under changing environment is very important since advanced applications become pervasive and ubiquitous, and need to adaptive to their changing context. So we design this classifier selection system using context-driven genetic algorithm. Classifier selection is a combination scheme that can be used in a classification system. This kind of system can tolerate the dynamic of varying environments. It adopts the context and comparability analysis for picking out an optimal classifier. The proposed scheme using genetic algorithm which is based on context analysis is suitable for the varying environments´ situations, we call this method context-driven genetic algorithm for short. The goal of this system is to select an optimal classifier from a group of candidate classifiers for the identified multiple possible clusters. It tries to distinguish the category of input environment and decides an optimal classifier.