Kaspersky fraud prevention cloud enables machine learning for multi-channel protection

Kaspersky Lab has released Kaspersky Fraud Prevention Cloud, a new solution for organizations facing risks from fraudulent activity via fast-growing online services. In addition to fraud prevention solutions for endpoints and mobile devices within Kaspersky Fraud Prevention platform, the new product features a set of cloud-based technologies designed to give banks, financial institutions, loyalty schemes providers and government agencies protection against fraudsters.

These include a global device reputation database, device and environmental analysis, behavioral analysis and biometrics, and clientless malware detection. With the rise of online and mobile banking, organizations need to fight fraud and money laundering while also ensuring protection for their users. For example, one in four customers of banks have been a victim of financial fraud in the last year. The new fraud prevention offering from Kaspersky Lab delivers multi-channel protection for both organizations and users, resulting in reduced losses from fraud and controlled prevention costs.

The solution incorporates advanced technologies to improve the visibility and detection of suspicious activity without undermining the user experience. Behavioral analysis and biometrics help to identify whether a person is real, without any additional actions or procedures required by the user. Behavior is analyzed through mouse movements, clicks, scrolls, keystrokes on PCs, and accelerometer/gyroscope position and gestures (touch, swipes and etc.) on mobile devices.

Kaspersky Fraud Prevention Cloud accumulates and analyzes user behavior, device, environment and session information as anonymized and depersonalized big data in the cloud, making it available to expert forensics and automatic offline analysis. This new information feeds into an organizations internal Enterprise Fraud Management system, which enables proactive fraud detection in real time, even before a transaction occurs. This approach is based on Humachine intelligence by Kaspersky Lab — a combination of big data and threat research analysis with machine learning algorithms and the expertise of the company’s best security teams.

Risk Based Authentication (RBA) assesses the risks before a user is logged into a digital channel, providing decisions to internal back-end systems on whether to proceed, request additional authentication information or block access until further verification. This feature improves usability for ‘legitimate users’ by decreasing the number of authentication stages, while the ‘unauthorized users’ are detected before they commit any fraudulent activity.

Continuous Session Anomaly Detection also helps to maximize fraudulent detection by identifying account takeover, new account fraud, money laundering, automated tools or any suspicious processes that occur during the session. As such, Kaspersky Fraud Prevention Cloud acts not only during the login process, but also during the whole session, building statistical models of various behavioral patterns with the help of machine learning technologies.

Clientless malware detection as part of Kaspersky Fraud Prevention Cloud combines direct and proactive detection techniques. The first identifies whether a customer’s device is used to directly attack a particular organization’s digital services. The second helps to identify malware that is not affecting the organization directly but may potentially be adapted for this purpose in the future. This helps a company to minimize risks and avoid losses of an actual attack when one occurs.