1st Open Call
Statice will solve the Cashier Fraud Detection challenge by Migros. Cashier and customer fraud can amount to large losses, given the huge number of transactions they deal with everyday. Being able to detect fraudulent transactions in real-time allows companies to make significant savings. But this can be tricky, especially in the case of employee or customers sensitive data. With stricter global privacy regulations in place, it is not sufficient anymore for companies to just pseudonymise data. Regulations such as the GDPR explicitly state the need for truly anonymous data.
In the past, it has been shown that pseudonymised data is not enough to guarantee the anonymity of the individuals in the datasets. Statice will show that it is possible to build highly performant solutions such as fraud detection on truly anonymous data by using synthetic data.
Statice is an automatic data anonymization software that generates entirely anonymous synthetic data. This synthetic data preserves the structure and utility present in the original data. Unlike traditional anonymization technologies and products, Statice enables highly complex secondary data use cases such as building performant machine learning models on anonymous data. It ensures that no personal information is ever exposed by providing privacy-preserving synthetic data which allows companies to run meaningful data analysis on a new synthetic dataset without ever touching original data.