Latest research in Criminal Network Analysis using Machine Learning focuses on employing advanced algorithms, including Graph Neural Networks (GNNs) and Graph Attention Networks (GATs), to uncover hidden structures, detect communities, and predict linkages within criminal networks. Studies aim to identify key players, understand hierarchical relationships, and forecast criminal activities by analyzing complex relational data. Techniques such as link prediction, community detection, and attention mechanisms are used to model dynamic network interactions and prioritize significant connections. These approaches enhance law enforcement capabilities by providing actionable insights into organized crime, optimizing resource allocation, and supporting proactive strategies for crime prevention and investigation.