Fraud Detection for Account Impersonation

Fraudsters are constantly looking for opportunities to prey on victims for financial gains. It is no surprise that during unexpected disruptions such as the pandemic, the fraudulent activities dramatically increase with businesses facing significant financial losses.

A Fortune 500 company has a requirement to develop a system to effectively monitor for account impersonation and takeover activity. The company’s goal is to proactively prevent any high-risk sessions from proceeding or for fraudulent payment transactions from being monetized.

GoodLabs, with its expertise in machine learning and real-time data analytics, was invited by the company to take on the challenge. GoodLabs’ engineering team has built a real-time fraud detection foundation framework that is capable of feeding third-party digital identity intelligence and advanced behavioural insights data sources into a massively scalable real-time and large scale analytics data lake. This allows the company’s fraud analytics team to slice and dice massive amounts of data in a much shorter time frame. GoodLabs’ A.I. team also applied algorithms such as isolation forest and autoencoders to train on normal behaviors to detect anomalies to reduce the company’s fraud analytics team to crawl through terabytes of data manually.

GoodLabs, together with our client, builds a better and safer tomorrow with advanced software engineering.