Virtual indicators of sex trafficking to identify potential victims in online advertisements

A content analysis of online sex worker advertisements suggests specific terms, sources and patterns of behavior that may help identify potential sex trafficked victims within these virtual environments. While some ads are posted by independent sex workers, others may have been posted by traffickers or pimps, advertising the women they have under their control. A total of 600 ads from the Backpage “escort” services site were harvested and analyzed for information elements including location, age, name, area code, ethnicity and controlled/restricted movement. The results suggest that 75% of the ads contained one or more of the primary sex trafficking indicators, and 15% of the ads contained three or more indicators. The results suggest that some physical indicators should be adapted to the virtual environment, and that the presence of multiple indicators can be used to prioritize certain ads for further investigation by law enforcement. It is hoped that these indicators can inform the design of future systems to flag high-risk advertisements, and help identify and disrupt this covert network activity.