Retail provider detected 70 % of known fraud cases and additional unexpected fraud cases
We detected shop assistants who commit fraud by stealin customer money.
We developed a detector which process surveillance CCTV data, pair it with the data from points of sale and detect frauds.
Platform is able to process videos from tens of shops in a batch over a night. The result is a detection of about 70 % of existing cases.
Area: Computer vision, safety & security
Case Studies
Automatic categorization for 98.5% of card transactions
With millions of clients conducting millions of operations every day,...
Just-in-time loan offers: a 10x higher conversion rate
Together with the bank, we used several years’ worth of...
Get inspired on our blog
Good data management and the gap between IT and business are the most burning issues for large Czech companies
Large Czech companies consider the biggest challenge they will face in the next three years to be good data management and business’ inability to...
Interview: Data literacy is growing, but there is still room for improvement, innovation, and automation. In the future, simplicity, intuitiveness, and speed will prevail, agree on Bínová and Bergner from ADASTRA
Adastra has been involved in data storage, processing, cleaning, analysis, and business use for over a quarter of a century. We asked our colleagues...
8 tips for building mature Big Data platforms
Companies who base their business on data generate more revenue than those who rely on intuition or other approaches in their decision-making. Don’t risk...