• Subscribe

Supercomputers expose the telltale signs of human traffickers

Jeff Schneider. Photo courtesy Carnegie Mellon University.

If you've ever read J.R.R. Tolkien's The Hobbit, you know every villain has a weakness. Tolkien's villain was the dragon Smaug - and, in his case, the weakness was a bare patch in his under-armor.

Like Smaug, our real-world villains have weaknesses we can exploit. For the villains running human trafficking operations, the chink in their armor is the need to attract customers. This insight secured a three-year, $3.6 million DARPA-Memex grant for Jeff Schneider, research professor in the Robotics Institute at Carnegie Mellon University's School of Computer Science in Pittsburgh, Pennsylvania, US.

The DARPA grant calls on CMU and partners to develop domain-specific indexing, so law enforcement agents can quickly sift through mountains of public internet data and find the telltale signs of human trafficking. Since traffickers typically disguise their movements, the grant will also be used to create machine-learning algorithms that identify specific language and image patterns used within that domain.

"Human traffickers have an Achilles heel because at the end of the day they need to advertise their service in a publicly accessible way, " Schneider observes. "That's the hook that lets us know there actually is content on the web that can help us locate them."

CMU had been working on an online human trafficking tool when DARPA called for grant applicants. Emily Kennedy, now CEO of Marinus Analytics, created a program called TrafficJam, while working on her CMU senior honors thesis.

"What prompted me to start this project was learning that while technology has empowered criminals to more efficiently exploit their victims, law enforcement doesn't have the time or the ability to develop the technology to respond," Kennedy says.

She demonstrated a prototype to local law enforcement agents, who were immediately impressed by the advantages they gained.

Emily Kennedy. Photo courtesy Carnegie Mellon University.

"One detective may have the experience and context to read through pieces of data online and know which ones would be relevant," Kennedy explains. "But since there are so many pieces that's not going to be feasible given time and resource constraints."

Marinus and CMU are but two parts of a larger partnership between academics, defense contractors, and law enforcement aiming to bring human trafficking to an end. "Many of the participants in this DARPA program are looking at cloud computing, and I think we'll do the same," says Schneider. Auton Labs will also lend 360 cores with 3.6 teraFLOPS of theoretical peak performance computing to hone the crime-fighting algorithms.

Ending human trafficking is a daunting task. The International Labour Organization (ILO) conservatively estimates 21 million victims of forced labor and 4.5 million victims of forced sexual exploitation worldwide.

"Of course, sex trafficking isn't the only criminal activity that occurs online," Schneider notes. He also sees a use for these tools in labor trafficking, drug smuggling, and money laundering. "I think eventually the tools we are working on will be appropriate for any domain in which someone wants to do better research, and it may be nothing to do with law enforcement or criminal activity."

For now, the victims of sex trafficking are continually exposed to danger, and provide limited connections to the criminal network that exploits them.

"The goal and mission of our project is to shine a light on those hiding in the shadows so that law enforcement can get the people who are behind these crimes, not just rescue the victims who are out in the open," Kennedy says.

Join the conversation

Do you have story ideas or something to contribute? Let us know!

Copyright © 2021 Science Node ™  |  Privacy Notice  |  Sitemap

Disclaimer: While Science Node ™ does its best to provide complete and up-to-date information, it does not warrant that the information is error-free and disclaims all liability with respect to results from the use of the information.


We encourage you to republish this article online and in print, it’s free under our creative commons attribution license, but please follow some simple guidelines:
  1. You have to credit our authors.
  2. You have to credit ScienceNode.org — where possible include our logo with a link back to the original article.
  3. You can simply run the first few lines of the article and then add: “Read the full article on ScienceNode.org” containing a link back to the original article.
  4. The easiest way to get the article on your site is to embed the code below.