With the explosive development of multi-view data from diverse sources, multi-view clustering (MVC) has drawn widespread attention. Existing MVC methods still have several limitations. First, it is difficult to sufficiently consider the local invariance within data views. Second, view fusion usually utilizes weighted averages, thus how to fuse views is warranting further exploration. Towards these two issues, this paper proposes a multi-channel augmented graph embedding convolutional network (MAGEC-Net) for multi-view clustering and its extended end-to-end model (EMAGEC-Net). The proposed frameworks are dedicated to exploring the consistency and complementarity of multi-view data. Specifically, on one hand, the augmented graphs are derived from generative adversarial networks, which explore the information and features of a single view more comprehensively. On the other hand, each augmented view is considered as a channel and fused by a deep fusion network, thus this method effectively improves the complementary information across views. Finally, feature extraction is performed on the fused consistent graphs to enable better clustering. Extensive experiments on six real challenging datasets demonstrate the effectiveness of the proposed method and its superiority over eight compared state-of-the-art methods.

Many studies in recent years have focused on social image understanding due to the increasing number of shared images from social networks and online communities. However, previous work in social image understanding fails to learn an effective feature representation because of a large amount of missing and irrelevant tags, though matrix completion techniques are frequently utilized for this purpose. Autoencoder models have been validated to be effective in learning latent low-dimensional representations in unsupervised learning. In this paper, we propose a new social image understanding model based on deep autoencoders, which can learn the shared latent codes of social images and tags as supervision information in the deep autoencoders. First, social images are extracted with multi-modal features, which provides a comprehensive characterization to image semantic understanding. And, the social image understanding problems are transformed into the problem of minimizing an optimization objective. Second, multi-layered autoencoders with weak supervision integration are employed to learn an efficient low-dimensional representation from the multi-view feature sources that can make up the semantic gap between image features and tags through minimizing the problem formulation. Finally, we design a new balanced loss function based on binary cross entropy, in which we address highly sparse inputs for a better optimization performance. Extensive experiments on several real-world social image datasets confirm the effectiveness and robustness of the proposed model compared with the state-of-the-art methods.