Robust Multi-Label Classification with Enhanced Global and Local Label Correlation

Data representation is of significant importance in minimizing multi-label ambiguity. While most researchers intensively investigate label correlation, the research on enhancing model robustness preliminary. Low-quality data one main reasons that degrades. Aiming at cases with noisy features and missing labels, we develop a novel method called robust global local correlation (RGLC). In this model, subspace learning reconstructs intrinsic latent immune from feature noise. The manifold ensures outputs obtained by matrix factorization are similar low-rank if similar. We examine co-occurrence constructed labels. Extensive experiments demonstrate classification performance integrated information statistically superior over collection state-of-the-art approaches across numerous domains. Additionally, proposed shows promising when labels occur, demonstrating classification.