Search

Graph neural networks, as the promising methodology in data mining for graph data, currently attract much attention and are broadly applied in graph-based tasks. Existing GNN methods mostly follow the assumption of homophily, where the connected nodes are similar and share the same labels. Most graphs in the real world can satisfy the assumption. However, for the particular nodes, the situation is not always satisfied. The connections between different-labeled nodes will introduce noise in feature aggregation and result in node representation deviating in the wrong direction. In this paper, we focus on the different-labeled neighbors of labeled nodes in the graphs. By regarding aggregation among neighbors as the procedure of node feature reconstruction, we devise a novel metric neighbor consistency to measure the difference between nodes and their neighborhoods. In this way, we can evaluate the reliability of nodes after aggregation. Furthermore, we propose a novel method, Neighbor Consistent Graph Neural Networks (NC-GNN), to promote the training of graph neural networks by reweighting the influence of labeled nodes based on neighbor consistency scores. Systematic experiments are conducted on benchmark datasets, and the results demonstrate the effectiveness of our method.

How to classify incredible messages has attracted great attention from academic and industry nowadays. The recent work mainly focuses on one type of incredible messages (a.k.a rumors or fake news) and achieves some success to detect them. The existing problem is that incredible messages have different types on social media, and rumors or fake news cannot represent all incredible messages. Based on this, in the paper, we divide messages on social media into five types based on three dimensions of information evaluation metrics. And a novel method is proposed based on deep learning for classifying the five types of incredible messages on social media. More specifically, we use attention mechanism to obtain deep text semantic features and strengthen emotional semantics features, meanwhile, construct universal metadata as auxiliary features, concatenating them for incredible messages classification. A series of experiments on two representative real-world datasets demonstrate that the proposed method outperforms the state-of-the-art methods.

Graph Neural Networks (GNNs) are powerful tools for modeling graph-structured data to solve the tasks such as node classification, link prediction along with graph classification. For the graph classification task, properly defining the pooling strategies to vary the size and structure of the input graph, is of vital importance to generate a graph-level representation of the input graph. However, the existing GNN models usually fail to effectively capture the graph substructure information in pooling process. Besides, the importance of nodes(supernodes) within a graph has not been well-reflected. To remedy these limitations, we propose Gated Structure Aware Pooling (GSAPool), a sparse and differentiable pooling method, which focuses on retaining the graph substructure information during the process of pooling in an end-to-end fashion. Specifically, GSAPool utilizes dual gates along with a self-attention network to integrate the local structure to form clusters' embeddings. It also employs a novel formulation to capture the importance of each node/supernode in an input graph. Experiment results show that GSAPool achieves competitive graph classification performance over the state-of-the-art graph representation learning methods.