Verizon Communications Inc. is planning to beef up the security of its enterprise network products following its acquisition Friday of startup Niddel Corp., which offers a threat-hunting platform based on machine learning technology.

Terms of the deal were not disclosed, but Verizon said it plans to integrate Niddel’s technology with its own products and services soon.

Niddel’s main product is a subscription service called Niddel Magnet, which leverages machine learning to hunt for security threats in customers’ networks. It’s a reactive tool that works by identifying machines in the network that have been potentially infected or compromised in some way before throwing out alerts.

“Using machine learning to improve information accuracy significantly reduces false positives and significantly improves our detection and response capabilities,” Alexander Schlager, Verizon’s executive director for security services, said in a statement Friday.

Reducing the number of false positives is one of Niddel’s main goals, according to its website. “[Niddel Magnet] eliminates up to 96 percent of false positives, compared to traditional indicator-based threat hunting techniques. On average 40 percent of alerts represent novel threats identified by our patented supervised machine learning models’ extrapolation of existing threat knowledge,” the company says.

More specifically, Niddel Magnet works by investigating the relationship between indicators of compromise and the tactics, techniques and procedures attackers use to steal data. It analyzes data from more than 50 internal and external sources and is fully autonomous, which means customers don’t need any specific security expertise to get the tool up and running.

Verizon currently offers a number of network and gateway security, security monitoring and incident response tools, so it’s likely Niddel’s capabilities will be added to some of these products.

Image: Verizon