Scaling the Web: Unraveling Online Reptile Leather Trade Networks with Machine Learning and Network Analysis

• In Event: GREEN Panel 3, Disrupting Criminal Networks: From Waste to Wildlife

Abstract

Since COVID-19, the illegal wildlife trade (IWT) has made a massive transition from physical to online marketplaces, creating new challenges for identifying and tracking the trade of reptile leather. Social network analysis and network analysis disruption have been used within criminal justice and IWT research to identify networks of key actors and generate strategies to disrupt these networks. However, these analyses have been limited to actors interacting in the physical space and have yet to explore online marketplaces. We utilize machine learning (ML) and large language models (LLMs) to extract data on potential illegal sales of small leather items on eBay as the case-study marketplace. We use social network analysis to identify key actors, products, and eBay sites where these activities occur. We then use different network disruption strategies (i.e., random node removal vs sequential removal based on network centrality measures) to determine the most optimal impact of disruption on network dismantlement. Our findings can inform targeted and effective responses to IWT across online marketplaces. Limitations and policy implications are discussed.

Authors

• Joshua Aaron Lang, John Jay College of Criminal Justice | CUNY Graduate Center

• Gohar A. Petrossian, John Jay College of Criminal Justice

• Bryan Lieu, Macaulay Honors College | Baruch College CUNY

• Kevin Bernstein, Baruch College

• Ulhas Gondhali, John Jay College of Criminal Justice, City University of New York

• Juliana Silva Barbosa, New York University

• Juliana Freire, New York University

• Sunandan Chakraborty, Indiana University Indianapolis

• Kinshuk Sharma, Indiana University Indianapolis