Factors Influencing the Adoption of Machine Learning Algorithms to Detect Cyber Threats in the Banking Industry

Abstract

Cyber attacks have evolved, making predicting and preventing their occurrence difficult. The complexity of cyber threats has contributed to the development of technology-intensive security systems, but these methods have yet to eliminate cyber threats effectively. Machine learning algorithms have proven helpful in cybersecurity applications for organizations. Machine learning algorithms provide a great opportunity for organizations to counter the increasing cyber-attack threats. Adopting machine learning algorithms in cyber-security is advantageous for organizations as they are more effective, scalable, and actionable in detecting cyber threats like malware than conventional methods, which require human involvement. Despite the great potential of machine learning in detecting cyber threats, financial institutions have recorded a very low adoption rate of these technologies in their efforts to mitigate the increasing cyber threat in the banking sector. In this quantitative nonexperimental, predictive correlational study, the Unified Theory of Acceptance and Use of Technology (UTAUT) model was used to explore the factors influencing the adoption of machine learning algorithms for detecting cyber threats among information technology (IT) professionals in the banking industry. The results showed that performance expectancy and facilitating conditions correlate positively with IT professionals’ behavioral intention to use machine learning algorithms to detect cyber threats. Effort expectancy had no significant effect, while social influence negatively influenced IT professionals’ behavioral intention to use machine learning algorithms to detect cyber threats. Determining these factors is a significant step towards eradicating the growing cyber threats in the banking industry.