Label Enhancement with Sample Correlation via Sparse Representation

Xiaoqian Zeng, Qin Qin, Chao Tan, Genlin Ji

Abstract: Label distribution describes the instance by multiple labels with different intensities and accommodates to more-general conditions compared with single-label and multi-label annotations. In the past few years, many label enhancement (LE) algorithms are proposed in order to solve the problem that many training sets cannot use label distribution learning (LDL) algorithms because they only contain logical labels rather than label distribution. To handle this problem, this paper proposes a novel LE algorithm called Label Enhancement with sample correlations via Sparse Representation (LE-SR). Unlike most existing methods, a sparse representation method is employed so as to capture the global relationships of samples and predict implicit label correlations to achieve label enhancement. Experimental results on thirteen real-world datasets show clear advantages of LE-SR over several existing LE algorithms.