In January 2021, one paper of Associate Professor Tan Chao in the Data Mining Research Group of our shcool is accepted by IEEE Transactions on Knowledge and Data Engineering (TKDE), an international top-tier journal in the CCF A. The paper is titled A Novel Probabilistic Label Enhancement Algorithm for Multi-label Distribution Learning [1]

This paper proposes a novel probabilistic label enhancement algorithm, called PLEA, to solve the challenging label distribution learning (LDL) for multi-label classification problems. They use the well-known label distribution learner based on the maximum entropy model. However, unlike the existing LDL algorithm which bases on the maximum entropy model, they use manifold learning to enhance the label distribution learner. The obtained experiment results show that the proposed PLEA method has advantages in LDL accuracy and runtime performance compared with the latest multi-label LDL method. The results also show that PLEA has advantages over the latest multi-label learning algorithms for classification tasks.