Uncovering the Darknet: A Hybrid Intelligence Framework for Drug Market Surveillance
Diogo Rio, Krishna Teja Atluri, Giuseppe Cascavilla, and Jessica De Pascale
To overcome the limitations of traditional drug monitoring systems, this work presents a hybrid intelligence tool that integrates data from darknet forums with datasets from international agencies to enhance drug landscape monitoring. This tool aims to aid law enforcement and public health agencies by providing actionable insights and trends over the years, as well as real-time developments in drug markets. NAFTA countries and the European Union served as the main geographical focus areas. The project is driven by data collected from d/Reviews from a darknet forum, Dread, where users post drug purchase experiences, offering up-to-date insights. Structured datasets complement this data on yearly summary of drug prevalence and pricing between 1999-2023 from the European Union Drugs Agency (EUDA) and the United Nations Office on Drugs and Crime (UNODC). Pre-processing techniques are applied on data to standardize and merge, enabling robust analyses such as time-series trend analysis, user-vendor network analysis, and text-based identification of emerging substances. These capabilities would allow stakeholders to identify threats more rapidly and respond proactively. Our end product demonstrates how unified data can support real-time exploration and decision-making. This hybrid approach addresses the growing need for dynamic, actionable intelligence in response to evolving drug trends.
