Peter Dornheim

Peter Dornheim

Auto-generated profile, based on publicly available information

Peter Dornheim

Institute of IT Management and Digitization (IFID), FOM University of Applied Sciences, Dusseldorf, Germany

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 October 2023 in Information & Computer Security
Verified
Journal Article
Open access
PDF file
PDPeter Dornheim
Peter Dornheim
RZRuediger Zarnekow
Ruediger Zarnekow
Purpose: The human factor is the most important defense asset against cyberattacks. To ensure that the human factor stays strong, a cybersecurity culture must be established and cultivated in a company to guide the attitudes and behaviors of employees. Many cybersecurity culture frameworks exist; however, their practical application is difficult. This paper aims to demonstrate how an established framework can be applied to determine and improve the cybersecurity culture of a company. Design/methodology/approach: Two surveys were conducted within eight months in the internal IT department of a global software company to analyze the cybersecurity culture and the applied improvement measures. Both surveys comprised the same 23 questions to measure cybersecurity culture according to six dimensions: cybersecurity accountability, cybersecurity commitment, cybersecurity necessity and importance, cybersecurity policy effectiveness, information usage perception and management buy-in. Findings: Results demonstrate that cybersecurity culture maturity can be determined and improved if accurate measures are derived from the results of the survey. The first survey showed potential for improving the dimensions of cybersecurity accountability, cybersecurity commitment and cybersecurity policy effectiveness, while the second survey proved that these dimensions have been improved. Originality/value: This paper proves that practical application of cybersecurity culture frameworks is possible if they are appropriately tailored to a given organization. In this regard, scientific research and practical application combine to offer real value to researchers and cybersecurity executives.

5 June 2024 in Computers & Security
Verified
Journal Article
Closed access
AZAnna Zanke
Anna Zanke
TWThorsten Weber
Thorsten Weber
PDPeter Dornheim
Peter Dornheim
MEMathias Engel
Mathias Engel

30 October 2024 in IEEE Access
Verified
Journal Article
Open access
HRHelena Richard
Helena Richard
PDPeter Dornheim
Peter Dornheim
TWThorsten Weber
Thorsten Weber
In the dynamic and competitive telecommunications industry, effective risk management is critical. Traditional methods are often manual and labor-intensive, prone to inaccuracies and inefficiencies. This paper addresses these challenges by integrating advanced machine learning algorithms to automate and improve the risk management process at a large telecommunications company, hereafter referred to as TP. The study applies state-of-the-art transformer-based machine learning models - BERT, RoBERTa, DeBERTa, and ERNIE 2.0 - alongside a classical support vector machine (SVM) for comparative analysis. The approach leverages the Cross-Industry Standard Process for Data Mining (CRISP-DM) model, emphasizes text classification algorithms to categorize risk events, and introduces a one-class SVM for novelty detection to identify unprecedented risk events. The results show that transformer-based models achieve superior prediction accuracy over traditional SVMs, with the best-performing transformer model (DeBERTa) achieving an F1 score of 92.5%, significantly improving the risk classification accuracy. In addition, the novelty detection mechanism successfully identifies novel risk events with an accuracy of 77.2% and an AUROC of 79.2%. This integration streamlines risk management processes and provides operational benefits to telecommunications companies. By providing insight into these technologies’ economic and operational benefits, this study contributes to the practical application of AI in telecommunications, particularly in risk event assessment and natural language processing.

27 September 2020
Verified
Book Chapter
Closed access
PDF file
PDPeter Dornheim
Peter Dornheim
RZRüdiger Zarnekow
Rüdiger Zarnekow