Online Banking Fraud Detection

There are two main points where the user can be targeted by fraudsters: during authentication (login) and during authorization (operation).

In the first scenario, the main goal of the attacker is to compromise user data, steal credentials or achieve an account take over (ATO). In the second scenario, the fraudster tries to bypass authorization mechanisms.

Several techniques can be deployed to achieve these goals: phishing, RAT, malware, webinjects, form-grabbers and others. Moreover, most of these attacks can be executed anytime during the user’s session.

bugFraud collects data throughout the whole session to prevent online fraud at every stage.

Machine Learning in Cyber Security to detect unknown threats

bugFraud uses a proprietary pattern similarity-based mechanism that can detect new malware campaigns that interact with the victim’s browser like MITB threats, regardless of whether or not the interaction matches a known malware signature. Our approach goes beyond the known signatures or blacklist, analyzing whether web page navigation has been subverted by fraudsters. buguroo® bugFraud Defense classifies grey anomalies to avoid false positives and false negatives.