一种非负稀疏近邻表示的多标签学习算法.pdfVIP

44 6 Vol.44 No.6 201511 Journal of University of Electronic Science and Technology of China Nov. 2015  ( 230601) k-LASSO k(ML-KNN)(ML-SRC) ; ; LASSO; TP391.4 A doi:10.3969/j.issn.1001-0548.2015.06.018 A Non-Negative Sparse Neighbor Representation for Multi-Label Learning Algorithm CHEN Si-bao, XU Dan-yang, and LUO Bin (School of Computer Science and Technology, Anhui University Hefei 230601) Abstract In order to avoid the influence of the nonlinear manifold structure in training data and preserve more discriminant information in the sparse representation based multi-label learning, a new multi-label learning algorithm based on non-negative sparse neighbor representation is proposed. First of all, the k-nearest neighbors among each class are found for the test sample. Secondly, based on non-negative the least absolute shrinkage and selectionator operator (LASSO)-type sparse minimization, the test sample is non-negative linearly reconstructed by the k-nearest neighbors. Then, the membership of each class for the test sample is calculated by using the reconstruction errors. Finally, the classification is performed by ranking these memberships. A fast iterative algorithm and its corresponding analysis of converging to global minimum are provided. Experimental results of multi-label classification on several public multi-label databases show that the proposed method achieves better performances than classical ML-SRC and ML-KNN. Key words LASSO sparse