A big data approach to risk management and control: Cybersecurity in accounting
Petar Halachev · Periodicals of Engineering and Natural Sciences (PEN) · 2024
This research reviews into the critical interplay between big data applications and cybersecurity risks within the accounting sector. Aimed at understanding how big data can mitigate these risks, the study develops a novel theoretical model using differential equations. This model, rooted in a thorough empirical approach, undergoes validation through logistic regression analysis of responses from 200 participants. The analysis particularly focuses on how demographic and socio-economic factors influence cybersecurity perceptions. Data Breach Consistency emerges as a key factor, evidenced by a coefficient of 1.204 and an odds ratio of 3.331, indicating a substantial link between the recognition of data breaches and increased cybersecurity concerns. Malware and Ransomware concerns demonstrate a notable impact, with a coefficient of 0.907 and an odds ratio of 2.477, underscoring the gravity of these threats. Results further highlight the mitigating influence of Big Data Mitigation on cybersecurity risks, marked by a coefficient of 0.491. The robustness of the model is affirmed by an Area Under the Curve (AUC) score of 0.843, attesting to its efficacy in predicting cybersecurity concerns. The findings highlight the vital role of big data in formulating effective cybersecurity strategies. This study not only contributes to the academic discourse on the intersection of big data and cybersecurity in accounting but also offers practical insights for enhanced decision-making and policy formulation in the evolving digital business environment.