Human Trafficking Demand Reduction Strategies through Network Analytics and Simulation Modeling
PI: Michelle Alvardo, Co-PI: Mostafa Reisi Gahrooei
Award Period: 01/16/2020-10/15/2020
Human trafficking, illegal trade in persons for forced labor or sexual exploitation, poses a significant challenge to the national and global safety and security. It has been estimated that around 12 to 24 millions of men, women, and children around the globe are victims of this activity. Recent efforts to employ data analytics tools to layout the supply-chain network of this illicit activity and consequently design intervention plans has been constrained by the lack of data and has been mainly focused on the supply side of the network. While these supply-side efforts show promising results for better understanding this illicit activity and should be continued, demand side of this global challenge has remained unexplored and requires significant attention. This project considers managing the demand side of human trafficking through awareness campaigns distributed over a social network environment. More specifically, we expect three major developments: First, we build a team of research investigators and stakeholders across multiple disciplines to gain a better perspective toward this illicit activity and its actors. Second, we will design an agent-based simulation model that probabilistically and dynamically models the reaction of the social network users to the advertisements sent by traffickers and to the awareness campaigns sent by educators from government agencies or non-profit organizations. Finally, we will design a reinforcement learning framework to design a strategy for propagating awareness campaigns over a social network to minimize the response to trafficking advertisements, while adhering to resource constraints.