acm-header
Sign In

Communications of the ACM

ACM TechNews

Cybersecurity Researchers Build Better 'Canary Trap'


View as: Print Mobile App Share:
A canary.

Canary traps can be used to sniff out information leaks or, as in World Ware II, to create distractions that hide valuable information.

Credit: GCN.com

The WE-FORGE data protection system developed by Dartmouth College cybersecurity researchers uses an artificial intelligence version of the "canary trap," in which multiple false documents are distributed to conceal secrets.

The system uses natural language processing to automatically generate false documents to protect intellectual property.

WE-FORGE also adds an element of randomness to keep adversaries from easily identifying actual documents.

The algorithm computes similarities between concepts in a document, analyzes each word's relevance, then sorts concepts into "bins" and computes a feasible candidate for each group.

Dartmouth's V.S. Subrahmanian said, "The system produces documents that are sufficiently similar to the original to be plausible, but sufficiently different to be incorrect."

From Dartmouth News
View Full Article

 

Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account