Science & Engineering

USC Tool Taps Big Data to Bring Down Child Trafficking

• By Dian Schaffhauser
• 06/01/15

The average age of entry into prostitution in the United States is 14. The kind of profit a pimp can expect to make on a child prostitute each year is $150,000. Child trafficking operations spend about $45 million a year advertising their services on literally thousands of sites and millions of pages. A three-year research project at the University of Southern California wants to shutter those operations, and it's using big data searching to do it.

Two computer science professors, Pedro Szekely and Craig Knoblock, working out of the Information Sciences Institute, have developed a new search tool intended to use the Internet to turn advertising against human traffickers.

The work is being funded by Memex, a Defense Advanced Research Projects Agency (DARPA) program aimed at developing the next generation of search technologies and transforming the way they discover, organize and present results.

The researchers are treating their search project as a big data problem. The tool they've created combs through escort ads; downloads all relevant pages, including the ones those ads link to; discovers connections, folds the data into a repository; and provides query and analysis functionality to enable searching by law enforcement users.

DIG (for "Domain-specific Insight Graphs") lets users who are searching for a missing child believed to be trapped in the escort industry to search by phone number, location, alias and photo and recommend a way to reach them. Currently, the database contains content from 50 million Web pages and 2 billion records; it's growing at the rate of about 5,000 Web pages per hour.

"As the database continues to grow, DIG will be able to uncover new connections and patterns in the data, making it even more useful," said Knoblock, the director of information integration at the institute, in a prepared statement.

All of the code for DIG is open source and made available freely to law enforcement agencies. The project leaders expect to upgrade that quarterly over the course of the project, which began six months ago.

Eventually, they said, Szekely and Knoblock hope to enhance DIG to be able to flag potential victims and identify trafficking rings through their ads.

This isn't USC's first entry into the plight of human trafficking. The university sponsors the Technology and Human Trafficking Initiative from the Center on Communication Leadership & Policy. That project is studying the exploitation of modern communications technology in modern slavery.