Tech Fights Crime: The Ways Algorithms and Websites Police for Child Exploiters and Criminals

The internet is our 21st Century frontier, and with that wilderness comes its share of exploiters and criminals.

Among the more abominable offenders are child sexual abusers and traffickers.

But police and websites increasingly leverage technology's endless innovations to fight back.

Here's how Big Tech is advancing that detective work on many fronts, according to recent and ongoing research culled from the Computer Society Digital Library. (An abstract is free, in front of the paywall. For the full-length report, most of the research listed below is behind the paywall and requires a subscription.)

Dropbox, Microsoft launch digital investigations in the cloud

As more of our personal information becomes exposed to cloud computing and the Internet of Things, cybercriminals exploit opportunities to steal that data and wreak havoc on you or a business: The hackers engage in hacking/phishing or malicious attacks, stalk people, or even engage in the abuse and sexual exploitation of children.

But companies are fighting back.

Dropbox has implemented child abuse material detection software similar to Microsoft’s PhotoDNA, which identifies exploitative pictures on their servers and cloud storage products.

Other cybercrimes, like denial-of-service attacks—which Amazon experienced with its Elastic Compute Cloud (EC2) service—are managed by DDoS protection systems, say researchers Kim-Kwang Raymond Choo of the University of Texas at San Antonio; Christian Esposito of the University of Naples “Federico II”, Italy; Aniello Castiglione of the University of Salerno (Italy).

Read "Evidence and Forensics in the Cloud: Challenges and Future Research Directions" research (login may be required for full text)

Algorithm engineers join fight and assist Facebook, Microsoft

Mitali Thakor of Northwestern University knows the history of forensic science about child exploitation.

“The detection and removal of child pornography online is a vexing process, weighted with moral urgency,” she writes.

Computing experts are developing even better algorithms to scour the internet quickly for images and video of children being raped or photographed for the purposes of being trafficked or exploited.

The goal is threefold: to find children who are missing, to gather evidence against criminals, and to get the files removed from the internet.

These algorithms are being used by service providers like Facebook and Microsoft to report suspected child abuse images to the NCMEC, as required by the PROTECT Our Children Act of 2008, which Congress passed to provide greater resources for fighting child exploitation and trafficking.

Read "How to Look: Apprehension, Forensic Craft, and the Classification of Child Exploitation Images" research here (login may be required for full text)

A new tool called NuDetective is deployed

Brazilian authorities developed an algorithmic technology called the NuDetective Forensic Tool to identify child pornography files, wrote Mateus de Castro Polastro and Pedro Monteiro da Silva Eleuterio of the Brazilian Federal Police.

It's not perfect, but it's highly effective in doing an image analysis as human nudity or not and in doing a file name analysis for suspicious wording of child pornography.

"Although tests have shown the existence of false

positives, that is not really a problem since the NuDetective

was able to reduce dramatically the files to be analyzed on a

hard disk drive (from 330,595 to only 183 suspicious files on

second experiment)," the authors write.

"With low rates of false negatives and short

processing time, the Tool proved very effective to help

combat the sexual exploitation of children and adolescents.

Future developments include implementation of face and

texture detection algorithms to reduce the number of false

positives," they continue.

Read "NuDetective: A Forensic Tool to Help Combat Child Pornography through Automatic Nudity Detection" research (login may be required for full text)

Related crime-fighting tech research from the Computer Society Digital Library

Login may be required for full text.

• Crime data mining: a general framework and some examples
• A Novel Skin Tone Detection Algorithm for Contraband Image Analysis
• A Statistical Approach for Identifying Videos of Child Pornography at Crime Scenes
• Efficient Tagging of Remote Peers During Child Pornography Investigations
• Behavioural Evidence Analysis Applied to Digital Forensics: An Empirical Analysis of Child Pornography Cases Using P2P Networks
• Challenges of automating the detection of paedophile activity on the Internet
• Unintended Consequences: Digital Evidence in Our Legal System

About Michael Martinez

Michael Martinez, the editor of the Computer Society's website and its social media, has covered technology as well as global events while on the staff at CNN, Tribune Co. (based at the Los Angeles Times), and the Washington Post. He welcomes email feedback, and you can also follow him on LinkedIn.

About Lori Cameron

Lori Cameron is a Senior Writer for the IEEE Computer Society and currently writes regular features for Computer magazine, Computing Edge, and the Computing Now and Magazine Roundup websites. Contact her at l.cameron@computer.org. Follow her on LinkedIn.