Abstract
Ensemble learning is a method of combining learners, however the ensemble sizes are sometimes unnecessarily large which causes extra memory usage and decrease in effectiveness. Error Correcting Output Code (ECOC) is one of the well known ensemble techniques for multiclass classification which combines the outputs of binary base learners to predict the classes for multiclass data. We formulate ECOC for ensemble selection problem by using difference of convex functions (dc) programming and zero norm approximation to cardinality constraint. Experiments show that it outperforms the standard ECOC.