Noise-robust classification with hypergraph neural network

<p>This paper presents a novel version of hypergraph neural network method. This method is utilized to solve the noisy label learning problem. First, we apply PCA dimensional reduction technique feature matrices image datasets in order reduce “noise” and redundant features runtime constructing Then, classic graph based semisupervised method, semi-supervised network, our proposed are employed The accuracies these five methods evaluated compared. Experimental results show that achieve best performance when noise level increases. Moreover, at least as good network.</p>