17 August 2023 in IEEE Transactions on Computational Social Systems
In recent years, the increasing prevalence of mobile social network fraud has led to significant distress and depletion of personal and social wealth, resulting in considerable economic harm. Graph neural networks (GNNs) have emerged as a popular approach to tackle this issue. However, the challenge of graph imbalance, which can greatly impede the effectiveness of GNN-based fraud detection methods, has received little attention in prior research. Thus, we are going to present a novel cost-sensitive graph neural network (CSGNN) in this article. Initially, reinforcement learning is utilized to train a suitable sampling threshold, followed by neighbor sampling based on node similarity, which helps to alleviate the graph imbalance issue preliminarily. Subsequently, message aggregation is executed on the sampled graph using GNN to obtain node embeddings. Concurrently, the optimization objective for the cost matrix is formulated using the sample histogram matrix, scatter matrix, and confusion matrix. The cost matrix and GNN are collaboratively optimized through the backpropagation algorithm. Ultimately, the derived cost-sensitive node embedding is employed for fraudulent node detection. Furthermore, this study provides a theoretical demonstration of the effectiveness of adaptive cost-sensitive learning in GNN. Extensive experiments are carried out on two publicly accessible real-world mobile network fraud datasets, revealing that the proposed CSGNN effectively addresses the graph imbalance issue while outperforming state-of-the-art algorithms in detection performance. The CSGNN code and datasets can be accessed at https://github.com/xxhu94/CSGNN.
11 January 2024 in IEEE Transactions on Big Data
Along with the rapid evolution of mobile communication technologies, such as 5G, there has been a significant increase in telecom fraud, which severely dissipates individual fortune and social wealth. In recent years, graph mining techniques are gradually becoming a mainstream solution for detecting telecom fraud. However, the graph imbalance problem, caused by the Pareto principle, brings severe challenges to graph data mining. This emerging and complex issue has received limited attention in prior research. In this paper, we propose a G raph AT tention network with CO st-sensitive BO osting (GAT-COBO) for the graph imbalance problem. First, we design a GAT-based base classifier to learn the embeddings of all nodes in the graph. Then, we feed the embeddings into a well-designed cost-sensitive learner for imbalanced learning. Next, we update the weights according to the misclassification cost to make the model focus more on the minority class. Finally, we sum the node embeddings obtained by multiple cost-sensitive learners to obtain a comprehensive node representation, which is used for the downstream anomaly detection task. Extensive experiments on two real-world telecom fraud detection datasets demonstrate that our proposed method is effective for the graph imbalance problem, outperforming the state-of-the-art GNNs and GNN-based fraud detectors. In addition, our model is also helpful for solving the widespread over-smoothing problem in GNNs. The GAT-COBO code and datasets are available at https://github.com/xxhu94/GAT-COBO.
6 February 2020 in IEEE Access
Link prediction in directed networks has always been a hot topic in many fields including network science, information system and data mining. Intuitively, once links are endowed with certain orientations, their reciprocate nature can potentially provide extra information for guiding link prediction. However, the role of reciprocal links in the formation of directed closure triads and their ability to enhance link prediction accuracy are not thoroughly investigated yet in existing works. In this paper, we first design an empirical test to investigate the role of reciprocal links in different types of directed networks by taking advantage of null models. Subsequently, based on solid evidence of the empirical test, two novel weighting mechanisms for link prediction are proposed utilizing reciprocity as extra information. The performance of proposed methods is comprehensively studied on eight realistic networks compared with several groups of benchmarks. Experimental results indicate that the proposed methods are more effective and robust than two state-of-the-art weighting methods and eight well-performing similarity indices.
1 January 2024 in Chinese Journal of Electronics
Link prediction utilizes accessible network information to complement or predict the network links. Similarity is an important prerequisite for link prediction which means links more likely occurs between two similar nodes. Existing methods utilize the similarity of nodes but neglect of network structure. However the link direction leads to a far more complex structure and contains more information useful than the undirected networks. Most classic methods are difficult to depict the distribution of the network structure with incidental direction so the similarity characteristics of the network structure itself are lost. In this respect, a new method of local structure entropy is proposed to depict the directed structural distribution characteristics, which can be used to evaluate the degree of local structural similarity of nodes and then applied to link prediction methods. Experimental results on 8 real directed networks show that this method is effective for both area under the receiver operating characteristic curve (AUC) and ranking-score measures, and improved predictive capacity of the baseline methodology.