Multi-view Learning for the Identification of Risky Users in Dynamic Social Networks

2025-03-25

AI Summary

This paper addresses the insufficient accuracy of risk user identification in dynamic social networks by proposing a fine-grained modeling approach that integrates multi-view learning with temporal graph neural networks. Unlike existing methods relying solely on static content or structural features, our approach explicitly models the joint temporal evolution of user attributes, interaction relationships, and belief/interest dynamics—thereby overcoming the limitations of static network assumptions. Technically, we are the first to jointly leverage Temporal Graph Convolutional Networks (T-GCN), multi-view feature alignment, and dynamic node embedding to enable unified learning over heterogeneous temporal signals. Evaluated on real-world datasets, our method achieves a 12.7% improvement in F1-score over static baselines and successfully identifies 73% of users who exhibit early-stage risk behavior transitions, demonstrating significantly enhanced temporal sensitivity and predictive foresight.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Planning, Routing, and Scheduling: Temporal Planning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Social Networks and Social Media: Social media analysis through the lenses of networks
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Technological progress in the last few decades has granted an increasing number of people access to social media platforms such as Facebook, X (formerly Twitter), and Instagram. Consequently, the potential risks associated with these services have also risen due to users exploiting these services for malicious purposes. The platforms have tools capable of detecting and blocking dangerous users, but they primarily focus on the content posted by users and usually overlook additional factors, such as the relationships among users. Another key aspect to consider is that users' beliefs and interests evolve over time. Therefore, a user who can be considered safe at one moment might later become malicious, and vice versa. This work describes a novel approach to node classification in temporal graphs, aimed at classifying users in social networks. The method was evaluated on a real-world scenario and was compared to a state-of-the-art system that treats the network as a static entity. Experiments showed that taking into account the temporal evolution of the network, in terms of node features and connections, is beneficial.

Problem

• Identify risky users in dynamic social networks
• Consider temporal evolution of user features and connections
• Improve classification by analyzing multi-view relationships

Innovation

• Multi-view learning for dynamic social networks
• Temporal graph-based node classification
• Considers evolving user features and connections