A Deep Learning Framework for Evaluating Dynamic Network Generative Models and Anomaly Detection
This addresses the challenge of assessing dynamic network models for applications like disease outbreaks and social influence, though it appears incremental as it builds on existing graph and temporal techniques.
The paper tackles the problem of evaluating generative models and detecting anomalies in dynamic networks, where traditional static methods are inadequate, by introducing DGSP-GCN, a deep learning framework that integrates graph convolutional networks with dynamic graph signal processing; results show it outperforms baselines with error rates like MSE of 0.0645 on real-world datasets.
Understanding dynamic systems like disease outbreaks, social influence, and information diffusion requires effective modeling of complex networks. Traditional evaluation methods for static networks often fall short when applied to temporal networks. This paper introduces DGSP-GCN (Dynamic Graph Similarity Prediction based on Graph Convolutional Network), a deep learning-based framework that integrates graph convolutional networks with dynamic graph signal processing techniques to provide a unified solution for evaluating generative models and detecting anomalies in dynamic networks. DGSP-GCN assesses how well a generated network snapshot matches the expected temporal evolution, incorporating an attention mechanism to improve embedding quality and capture dynamic structural changes. The approach was tested on five real-world datasets: WikiMath, Chickenpox, PedalMe, MontevideoBus, and MetraLa. Results show that DGSP-GCN outperforms baseline methods, such as time series regression and random similarity assignment, achieving the lowest error rates (MSE of 0.0645, MAE of 0.1781, RMSE of 0.2507). These findings highlight DGSP-GCN's effectiveness in evaluating and detecting anomalies in dynamic networks, offering valuable insights for network evolution and anomaly detection research.