Design of a Federated Learning System for IT Security: Towards Secure Human Resource Management

• Lisa Verlande Universität der Bundeswehr München
• Ulrike Lechner Universität der Bundeswehr München
• Steffi Rudel Universität der Bundeswehr München

Resumo

Federated learning is a new, decentralized type of machine learning in which the models, rather than the data, are shared to maintain privacy in machine learning. This paper aims to investigate the design of a federated learning system to increase IT security in Human Resource Management and, in particular, recruiting processes while complying with business needs and General Data Protection Regulation. We propose a federated learning system and a novel approach to identifying malware throughout the recruiting process. The combination of Design science, Service Science, and the reference modeling method guide our research design. This paper presents the results of the first design iterations with the identification of service design elements and the design of a recruiting process with a federated learning system inside.

Palavras-chave: IT security, Federated Learning, Human Resource Management, Design Science