Metaheuristic Ensemble Pruning via Greedy-based Optimization Selection

Ensemble selection is a crucial problem for ensemble learning (EL) to speed up the predictive model, reduce the storage space requirements and to further improve prediction accuracy. Diversity among individual predictors is widely recognized as a key factor to successful ensemble selection (ES), while the ultimate goal of ES is to improve its predictive accuracy and generalization of the ensemble. Motivated by the problems stated in previous, we have devised a novel hybrid layered based greedy ensemble reduction (HLGER) architecture to delete the predictor with lowest accuracy and diversity with evaluation function according to the diversity metrics. Experimental investigations are conducted based on benchmark time series data sets, support vectors regression algorithm utilized as base learner to generate homogeneous ensemble, HLGER uses locally weight ensemble (LWE) strategies to provide a final ensemble prediction. The experimental results demonstrate that, in comparison with benchmark ensemble pruning techniques, HLGER achieves significantly superior generalization performance.

Improving Remote Sensing Multiple Classification by Data and Ensemble Selection

In this article, margin theory is exploited to design better ensemble classifiers for remote sensing data. A semi-supervised version of the ensemble margin is at the core of this work. Some major challenges in ensemble learning are investigated using this paradigm in the difficult context of land cover classification: selecting the most informative instances to form an appropriate training set, and selecting the best ensemble members. The main contribution of this work lies in the explicit use of the ensemble margin as a decision method to select training data and base classifiers in an ensemble learning framework. The selection of training data is achieved through an innovative iterative guided bagging algorithm exploiting low-margin instances. The overall classification accuracy is improved by up to 3%, with more dramatic improvement in per-class accuracy (up to 12%). The selection of ensemble base classifiers is achieved by an ordering-based ensemble-selection algorithm relying on an original margin-based criterion that also targets low-margin instances. This method reduces the complexity (ensemble size under 30) but maintains performance.