AUTOMATIC LEARNING DETECTION OF ANOMALIES IN A SET OF BANKING TRANSACTIONS BY OPTIMIZING THE AVERAGE PRECISION

摘要

The invention relates to a method for the detection of anomalies in a set of payment transactions, consisting in: - establishing (E3) a meta-model formed of a set of models, each trained on a training set to determine a risk for each transaction to be an anomaly, said metamodel being established by the "gradient boosting" technique, so as to optimize a differentiable function expressing the average accuracy of said meta-model; submitting (E4) said set to said meta-model, in order to determine risks for each transaction of said set, and, - determining a subset of transactions corresponding to a risk greater than a determined threshold to provide a predetermined number of transactions in said subset.