Incremental Learning with Ensemble Based Svm Classifiers for Non-Stationary Environments

In most of the real world applications, data are collected over an extended period of time and the distribution underlying the data is likely to change by time. In recent years, a lot of methods have been proposed for effective learning in changing environments. Incremental learning algorithms can also be used for learning in changing environments. In this work, we proposed to use incremental learning with ensemble of SVM classifiers for classification problems in changing environment and evaluated the performance of the proposed method on changing environments. Simulation results of the proposed algorithm on different non-stationary environments show promising results.