Multi-label Classification Using Rough Sets

In multi-label classification, each instance may be associated with multiple labels simultaneously which is different from the traditional single-label classification where an instance is only associated with a single label. In this paper, we propose two types of approaches to deal with multi-label classification problem based on rough sets. The first type of approach is to transform the multi-label problem into one or more single-label problems and then use the classical rough set model to make decisions. The second type of approach is to extend the classical rough set model in order to handle multi-label dataset directly, where the new model considers the correlations among labels. The effectiveness of multi-label rough set model is presented by a series of experiments completed for two multi-label datasets.