<ul><li>Link prediction in dynamic networks is a key challenge in network science, involving inferring potential interactions and their changing strengths over time.</li><li>Traditional static network methods have limitations in capturing temporal dependencies and weight dynamics, while tensor-based methods like tensor wheel decomposition (TWD) offer a solution by representing dynamic networks as high-order tensors.</li><li>This study introduces a PID-controlled tensor wheel decomposition (PTWD) model that utilizes TWD's power to capture dynamic network features and integrates PID control principle for stable model parameter learning.</li><li>The PTWD model shows improved link prediction accuracy on real datasets, demonstrating its effectiveness compared to other models.</li></ul>

A PID-Controlled Tensor Wheel Decomposition Model for Dynamic Link Prediction

Discover more