In this paper we investigate the use of fuzzy rule-based classifiers for multi-label classification. This classification task deals with problems where more than one label could be assigned simultaneously to a given instance. We concentrate on problem transformation methods, which use different strategies to transform a multi-label problem into a different single-label classification problems. This transformation make it possible to use almost any single label learner as base-classifiers, thus benefiting from the rich miscellany of algorithms available for this task. Fuzzy rules provide both interpretability and flexibility to model the vagueness among different labels. Empirical results using six datasets, four different problem transformation methods, eight base-classifiers, and five different performance measure shows the suitability of fuzzy rules for this task.