Abstract
The paper is devoted to the issues of controlling the operational risks of credit organizations associated with employees. One of the leading sources of operational risks associated with the actions of staff is insufficient employee qualification. This can cause a reduction in the accessibility and quality of the services offered by credit organizations, as well as potential financial and reputational losses. The goal of the study is to create an artificial neural network using a high-level Keras library in Python that would automatically monitor the level of critical personnel competence given its impact on the occurrence of operational risk events. The research objectives encompass analyzing employee competence indicators’ influence on operational risk occurrence, identifying key indicators for an artificial neural network, determining its architecture, creating datasets, and conducting comparative analysis of trained neural networks. The article outlines the primary set of indicators producing the greatest impact on the possibility of the emergence of operational risk associated with the actions of the credit organization’s employees. The paper also presents the results of testing the generated sets of training and test data using application software packages that implement mathematical methods to assess the consistency of the generated data sets. Graphs showing the results of training and testing of the constructed neural network are provided. The research findings are novel and can enable credit organizations to majorly automate the monitoring of personnel-related operational risk.
Keywords: Artificial Neural Network, Forward Propagation, Keras High-Level Library, Machine Learning, Neural Network.