Multi-Label Classification Methods for Multi-Target Regression

Real world prediction problems often involve the simultaneous prediction of multiple target variables using the same set of predictive variables. When the target variables are binary, the prediction task is called multi-label classification while when the target variables are real-valued the task is called multi-target regression. Although multi-label classification can be seen as a specific case of multi-target regression, the recent advances in this field motivate a study of whether newer state-of-the-art algorithms developed for multi-label classification are applicable and equally successful in the domain of multi-target regression. In this paper we introduce two novel algorithms for multi-target regression, multi-target regressor stacking and regressor chains, inspired by two popular and successful multi-label classification approaches. Furthermore, we develop an extension of the regressor chains algorithm which aims at improving its predictive performance and which is also applicable in the classification domain. All methods are empirically evaluated on 6 multi-target regression data sets, 4 of which are firstly introduced in this paper. The results of the evaluation show that all the proposed multi-target methods are able to improve the accuracy of the baseline approach which performs a separate regression for each target and that the corrected regressor chains method achieves the best overall accuracy.