Abstract: Aiming at the problem that autoencoder  only extracted features from  the content information contained in a single data, ignoring 
the structure information of data, we proposed a deep graph clustering network based on heterogeneous fusion and discriminant loss. Firstly, the heterogeneous information obtained by two autoencoders was fused, and the problem of information loss was solved when a single autoencoder was used to extract features. Secondly, the discriminant loss function was designed in the clustering training module based on the consistency of distribution within the same cluster, so that the model could be trained end-to-end, and avoiding the mismatch between the feature extraction and the assumptions of the clustering algorithm in the two-stage training methods. Finally, experiments were carried out on six commonly used datasets to verify the effectiveness of the proposed method. The experimental results show that compared with most existing deep graph clustering models, the proposed method  significantly improves the clustering performance on both non-graph and graph datasets.

Key words: graph clustering, deep learning, discriminant loss, heterogeneous fusion