An efficient stacking model with label selection for multi-label classification

Binary relevance (BR) is one of the most popular frameworks in multi-label learning. It constructs a group of binary classifiers, one for each label. BR is a simple and intuitive way to deal with multi-label problem, but fails to utilize label correlations. To deal with this problem, dependent binary relevance (DBR) and other works employ stacking learning paradigm for BR, in which all labels are viewed as additional features. Those works may be suboptimal as each label has its own most related label subset. In this paper, a novel two-layer stacking based approach, which is named a Stacking Model with Label Selection (SMLS), is induced to exploit proper label correlations for improving the performance of DBR. At the first layer, we construct several binary classifiers in the way of BR. At the second layer, we find the specific label subset through label selection for each labels , and expand them into feature space. The final binary classifiers are constructed based on their corresponding augmented feature space. Comprehensive experiments are conducted on a collection of benchmark data sets. Comparison results with the state-of-the-art approaches validate the competitive performance of our proposed approach. Comparison results with DBR shows that our approach is not only more time efficient but also more robust.