Ocado Technology (the technology division of Ocado, the world's largest online only supermarket) has revealed the world's first AI-based fraud detection system for online grocery purchases. The new system relies on an advanced machine-learning algorithm developed by Ocado Technology using TensorFlow and the power of Google Cloud.

 

The system is part of the Ocado Smart Platform (OSP), an end-to-end solution designed to power all aspects of a retailer's e-commerce business. OSP has been licensed by multiple grocers worldwide, including Morrisons in the UK, Groupe Casino in France, and Sobeys in Canada.

 

Predicting and recognising fraud

 

Online fraud typically covers any instance where an order is delivered but not paid for. Fraud can happen as a result of a genuine mistake (a customer entering the wrong personal details or using an expired card accidentally) –however, occasionally it can also be the result of malicious intent. If left unchecked, fraud can propagate to other systems and companies, and can affect customer service. 

 

In response to the challenge, Ocado Technology developed a mechanism of predicting and recognising those incidents among millions of other normal events using data collected from past orders, including cases of fraud. The engineers then implemented a deep neural network using TensorFlow and uploaded the whole fraud detection system to the cloud. 

 

The model has already achieved significant success, improving Ocado’s precision of detecting fraud by a factor of 15.