LIFT: Multi-Label Learning with Label-Specific Features

Multi-label learning deals with the problem where each training example is represented by a single in-stance while associated with a set of class labels. For an unseen example, existing approaches choose to determine the membership of each possible class label to it based on identical feature set, i.e. the very instance representation of the unseen exam-ple is employed in the discrimination processes of all labels. However, this commonly-used strat-egy might be suboptimal as different class labels usually carry specific characteristics of their own, and it could be beneficial to exploit different fea-ture sets for the discrimination of different labels. Based on the above reflection, we propose a new strategy to multi-label learning by leveraging label-specific features, where a simple yet effective algo-rithm named LIFT is presented. Briefly, LIFT con-structs features specific to each label by conducting clustering analysis on its positive and negative in-stances, and then performs training and testing by querying the clustering results. Extensive exper-iments across sixteen diversified data sets clearly validate the superiority of LIFT against other well-established multi-label learning algorithms.