Trajectory-FairBoost++: A Mobility-Aware and Fairness-Driven Framework for Influence Maximization in Multiplex Networks

Influence Maximization (IM) in complex networks is a foundational problem with applications spanning social media marketing, epidemiology, and information dissemination. While classical solutions assume static, single-layer networks and optimize solely for influence spread, real-world systems are inherently dynamic and multiplex, with users exhibiting cross-layer mobility. Moreover, existing models often neglect fairness, leading to biased seed selection across over-represented communities. To address these limitations, we propose Trajectory-FairBoost++, a novel framework that jointly optimizes for influence spread and community fairness in multiplex networks. Our method introduces a hybrid node scoring mechanism that combines mobility-aware trajectory frequency and bridge centrality, ensuring cross-layer influence propagation. To enforce fairness, we apply entropy-based community-aware selection to promote diverse seed representation. We further propose FairSpread++, a unified metric that integrates influence spread and fairness for rigorous evaluation. Experiments on real-world multilayer datasets under multiple diffusion models demonstrate that Trajectory-FairBoost++ consistently outperforms state-of-the-art methods in both efficiency and equitable influence coverage.