Authors: Hao Wang, Andrew Philpot, Eduard H. Hovy

Women and children are trafficked between countries and within countries for illicit sexual purposes. This is a serious international crime.

Domestic traffickers use a variety of means to advertise the illicit sexual services of the children and women they offer, including Internet classified ads, bulletin boards, and social media associated with escort and massage services (EMS). Clients (“johns”) of the EMS fronts for prostitution also use the Internet and social media to compare their experiences and offer leads to one another.

FBI, the principal U.S. law enforcement authority in this area, has begun implementing a number of initiatives to combat child sexual trafficking. We describe a prototype law enforcement support system to automatically compile and correlate information from open sources about trafficking and sexual abuse of women and especially children.

The system, called TrafficBot, employs information retrieval, information integration, and natural language technologies to build a data warehouse allowing various visualizations of information for the benefit of law enforcement. We discuss the current capabilities of TrafficBot, how it could be used by law enforcement, and suggest some future directions.