*5.1.8 Ensemble of classifier chains (ECC)*

This technique [23] uses classifier chains as a base classifier. It trains several CC classifiers using a standard bagging scheme. The produced binary models of each chain are ordered according to a random seed. Each model predicts different label sets. These predictions are summed per label so that each label receives a number of votes. A threshold is used to select the most popular labels that form the final predicted multi-label set.
