Data Ninja Labelling requires input data in form of a list of bank transactions (such as data obtained via open-banking AISP Apps) which is then analysed and labelled into one of 27 transaction categories with further sub-categories. Those categories cover a wide range of economic activity and ML engine, thanks to being trained on over 60,000,000 datasets, can precisely label any given transaction to the point of differentiating shopping for clothes from shopping for sports accessories. This data can later be used to fuel credit scoring models, predicting whether your users will make a purchase at your store or create their behavioural profiles for more precise marketing campaigns.