On Pinterest and Yelp, specialized machine learning algorithms scan user activity for signs of malicious behavior such as deceptive review practices. The algorithms are powered by a system from a Mountain View, California-based startup called DataVisor Inc. that today announced it has closed a $40 million round of funding.

Sequoia Capital’s Chinese arm led the investment with participation from New Enterprise Associates and GSR Ventures. The cash infusion marks the latest milestone in DataVisor’s growth streak. The startup claims to have helped protect more than 2 billion online users from fraud since it was founded by former Microsoft Corp. researchers Yinglian Xie and Fang Yu in 2013.

DataVisor’s fraud detection platform harnesses a type of artificial intelligence known as unsupervised machine learning to catch scammers. Whereas a regular machine learning model must be trained to recognize patterns with the help of carefully curated sample data, the unsupervised variety figures out what to look for on its own.

This is part of the reason DataVisor can detect not only traditional financial fraud but also other types of illicit activity. While Yelp and Pinterest rely on the startup’s platform to protect their services, an unnamed “top” app publisher employs the software to identify attempts to abuse promotions. A bank, in turn, could use DataVisor to protect client accounts from takeover attacks.

The startup said its platform can automatically identify new types of threats as they emerge. Moreover, the machine learning models under the hood look for correlations between seemingly disparate user activity to find scammer accounts that might otherwise go unnoticed.

The new Sequoia-led round will enable the startup to strengthen its position in the crowded fraud detection market. Although DataVisor’s pitch has won over some big-name customers, it faces tough competition from other providers that likewise make use of machine learning in their products. One such player is San Mateo-based Feedzai Inc., which raised a $50 million funding round late last year for its rival fraud prevention approach.

“Enterprises today are facing constantly evolving threats from sophisticated and tech-savvy fraudsters who continuously experiment and find ways to evade detection,” said DataVisor co-founder and Chief Executive Officer Yinglian Xie. “This new round of financing will enable us to further improve our technology, expand our services globally and transform the way that businesses fight fraud.”

Image: DataVisor