One-against-all ensemble for multiclass pattern classification

► A novel multiclass approach for multiclass pattern classification, abbreviated as OAAE, is proposed. ► OAAE incorporates a neural network ensemble into the one-against-all method to improve the generalization performance of the classifier. ► OAAE has successfully been tested on 20 multiclass pattern classification problems from UCI Machine Learning Repository. ► Results show that OAAE can reduce the uncertainty of the decision and it is comparable to the other widely used methods.