Interpretable models for the automated detection of human trafficking in illicit massage businesses

Abstract

Sexually oriented establishments across the United States often pose as massage businesses and force victim workers into a hybrid of sex and labor trafficking, simultaneously harming the legitimate massage industry. Stakeholders with varied goals and approaches to dismantling the illicit massage industry all report the need for multi-source data to clearly and transparently identify the worst offenders and highlight patterns in behaviors. We utilize findings from primary stakeholder interviews with law enforcement, regulatory bodies, legitimate massage practitioners, and subject-matter experts from nonprofit organizations to identify data sources and potential indicators of illicit massage businesses (IMBs). We focus our analysis on data from open sources in Texas and Florida including customer reviews and business data from Yelp.com, the U.S. Census, and GIS files such as truck stop, highway, and military base locations. We build two interpretable prediction models, risk scores and optimal decision trees, to determine the risk that a given massage establishment is an IMB. The proposed multi-source data-based approach and interpretable models can be used by stakeholders at all levels to save time and resources, serve victim-workers, and support well informed regulatory efforts.

Keywords:

• Human trafficking
• illicit massage
• interpretable machine learning
• risk scores
• decision trees

Data availability statement

This work utilizes several data sources with varying levels of accessibility to create one labeled data set for training and evaluating the proposed models. The geographic and socio-demographic data sets are publicly available. The licensing records obtained through public records requests cannot be shared but other researchers can submit the same public records requests. Access to the non-publicly available data sets, Yelp, Rubmaps, and advertisement phone numbers can be requested from Global Emancipation Network. Table A3 itemizes each data source used by the authors and explains how one can access or obtain the data set.

Additional information

Funding

This work was funded by the National Science Foundation CMMI Award #1936331. NSF;

Notes on contributors

Margaret Tobey

Margaret Tobey is a PhD student in the Operations Research Program at North Carolina State University. She received her bachelor’s degree in industrial and systems engineering in 2018 and master’s degree in operations research in 2020, both at NC State. Her research uses data science and machine learning to detect and disrupt human trafficking.

Ruoting Li

Ruoting Li is a PhD student in the Edward P. Fitts Department of Industrial and Systems Engineering at North Carolina State University. In 2018, she earned her bachelor’s degree in industrial and systems engineering from Lehigh University. Her research interests include developing interpretable machine learning models, with applications in healthcare and combating human trafficking.

Osman Y. Özaltın

Osman Özaltın is an associate professor in the Edward P. Fitts Department of Industrial and Systems Engineering at North Carolina State University. He is also a member of the Personalized Medicine Faculty Cluster. He received his MS and PhD degrees in industrial engineering from the University of Pittsburgh. His research interests span theoretical, computational, and applied aspects of mathematical programming, focusing on decision problems arising in personalized medical decision making and illicit supply chains. His methods include integer programming, combinatorial optimization, stochastic programming, bilevel programming, and decomposition algorithms for large-scale mathematical programs.

Maria E. Mayorga

Maria E. Mayorga is a professor of personalized medicine in the Edward P. Fitts Department of Industrial and Systems Engineering at North Carolina State University. She received her M.S. and PhD degrees in industrial engineering and operations research from the University of California, Berkeley. Her research interests include using mathematical programming, stochastic models and simulation for predictive models in health care, health care operations management, emergency response, and humanitarian logistics. In particular, she is interested in problems that help improve the human condition. In 2022 she became a Fellow of IISE.

Sherrie Caltagirone

Sherrie Caltagirone is the Founder and Executive Director of Global Emancipation Network (GEN), the leading data analytics and intelligence nonprofit dedicated to countering human trafficking. Prior to starting GEN, she served as a Policy Advisor for Orphan Secure, a global human trafficking rescue nonprofit, and began her anti-trafficking career with the Protection Project at the Johns Hopkins University. She received her degree in international relations summa cum laude from American University.