Security vendor Pindrop Security has added granular access controls for users and a new Fraud Workbench for call analysis to its phone Fraud Detection System (FDS), which helps identify potential phone fraud via a number of machine learning analytics techniques.
The new features are part of the 2.4 release of FDS, which was announced on July 28. Pindrop has raised $47 million in funding to date, with its most recent $42 million Series B million round announced on Feb. 19.
“The Fraud Detection System helps to detect fraudulent calls, but there had not been the ability to look at all the calls in an aggregated format,” Vijay Balasubramaniyan CEO, CTO and founder of Pindrop, told eWEEK.
That’s where the idea of the Fraud Workspace, which is debuting in FDS 2.4, comes into play. Balasubramaniyan explained that the Fraud Workspace will be useful for companies with their own fraud teams that want to look at the trends of fraud calls to identify specific patterns that attackers are using to try to deceive the company’s call center.
“Phone fraud is a cat and mouse game, with really smart people constantly changing attack techniques, which is why you need defenders staying two or three steps ahead,” Balasubramaniyan said.
Pindrop’s own development team used the Fraud Workspace tool to help build out the fingerprinting capabilities of FPS. With the FPS 2.4 milestone, the Fraud Workspace capability has matured and is being made publicly available to Pindrop’s customer base.
Balasubramaniyan explained that from a technology perspective, the core FPS database provides a real-time response for phone call fraud risk when a call comes in. For the Fraud Workspace, it’s a query-based system, which takes more time to determine a result.
“So you can ask the Fraud Workspace a bunch of questions, and it then goes through all the data,” Balasubramaniyan said.
If a user is able to identify a new trend with Fraud Workspace, that data can then be used to enhance the model for the real-time component of FPS.
The FPS 2.4 release also adds fine-grained access controls to help control and secure the data that the system collects and analyzes. In prior releases, Balasubramaniyan said, FPS only had simple access controls, since the FPS technology was only being deployed at individual lines of business units. He added that now Pindrop is getting much larger deployments and as such there is a need for more robust access control.
With fine-grained access control, for example, a company can specify that certain fraud analysts who are contractors not be able to look at certain calls that include confidential information, whereas other fraud analysts who are employees get access to all calls, Balasubramaniyan explained.
“So every element in the Pindrop tool can be turned on or off based on the access control role of the user,” he said.
The plan for future releases, Balasubramaniyan said, is to continue to integrate increasing levels of advanced analytics into the system. The roadmap is to turn the FPS into more of a platform with the introduction of an API that can enable the system to be integrated with other security and forensic tools.
Sean Michael Kerner is a senior editor at eWEEK and InternetNews.com. Follow him on Twitter @TechJournalist.
