Multi-dimensional classification (MDC) deals with the problem where an example is associated with multiple class variables that are assumed in the output space, and each class variable corresponds to one heterogeneous class space. Recently, label encoding has been proposed which transforms the original MDC output space into a new encoded label space to address the heterogeneity of class spaces. Nevertheless, the critical correlation information contained in features and labels was less taken into account. In this paper, we proposed a new approach called multi-dimensional classification using correlation information (MDCI). MDCI handles the heterogeneity of class spaces by utilizing one-hot label encoding to transform the original MDC output space into an encoded label space. Subsequently, a multi-dimensional classifier is induced based on sparse learning with \(\ell _{1}\) -norm regularization. Finally, the correlation information contained in features and labels is exploited to constrain the feature weight matrix to fit the information of feature and label better. Experimental comparisons on benchmark data sets demonstrate that our proposed method outperforms state-of-the-art MDC methods.