Ocado develops AI solution to fight retail fraud

Technology opens the door to endless opportunities for retailers, but with each new tool comes a unique set of security challenges. Whether the tech exists in a store or online, whether it takes the form of hardware or software, it’s a retailer’s duty to bolster security efforts to protect not just its business, but its customers too.

Last week, Ocado unveiled a new AI-powered system capable of detecting retail-sector fraud. As Computing.co.uk reports, the online supermarket has boosted detection rates on the cloud-based Ocado Smart Platform (OSP) “by a factor of 15” since its implementation.



The fraud detection system uses a machine learning algorithm developed by Ocado’s tech team, coupled with open-source software Tensorflow, and operates in Google Cloud. It will form part of the OSP, an end-to-end solution enabling retailers to digitise their business.

In a blog, the company acknowledged that fraud can happen due to a genuine mistake, such as a consumer entering incorrect personal details, but it can also be a result of malicious intent. “If left unchecked, fraud can propagate to other systems and companies and affect customer service,” the company wrote.

Ocado wanted to create a piece of technology that could use past purchase data to predict and recognise fraud incidents.

Holly Godwin, from Ocado Technology, said that machine learning is powerful for identifying fraud as it “can learn and adapt far quicker” than other technologies.

She continued: “The work of fraud agents is then made more manageable, as they no longer have to frantically [analyse] thousands of data points to establish fraud.

“Instead, they simply perform a final check to confirm whether they should cancel the order or not based on the prediction made by the model; it’s a perfect case of humans and machines working together in harmony.”



After collecting data from past purchases, like delivery information, software was used to implement a neural network into the system, and this was then sent to the cloud.

Godwin explained that her team put together a list of features which included the number of past deliveries and the cost of baskets, among other details. The more features, the more reliable the model, which is why the tech team added as much information as possible and will continue to add more in the future.

Though the model is in its early stages, Ocado said that its precision of detecting fraud has jumped by a factor of 15.

“We are now tackling our next challenges: investigating algorithms that could allow us to explain our predictions in more detail, assessing whether we can transfer learnings from one retailer to another, and considering what tools could help us to streamline our process,” Godwin said.

James Donkin, Ocado Technology’s general manager, said the system was created to tackle fraud first and foremost, but it also helps to enhance customer service.

“This is particularly important because fraud doesn’t affect only the retailer, it can also propagate to payment companies and other systems. Therefore, detecting and addressing fraud early is a win for the customer and for the retailer.”

We see embracing cloud technology as essential for success in today’s, competitive, digital-first landscape. Part of our recent transformation involved moving our website and other core services to the cloud, and we need people who possess the skills and know-how to make sure we stay on top of developments in this field. If you’re interested in joining our growing team, take a look at our current vacancies today.




Post Comment