Explainable Machine Learning in Cybersecurity

This study addresses the need for transparency in machine learning models within cybersecurity, emphasizing the importance of comprehensible explanations to support trust and informed decision-making. It proposes a framework to enhance explanation in two critical areas: identifying suspicious cryptocurrency transactions to improve information-level security and examining Android software for malicious behavior to strengthen system-level security. By clarifying the decision-making processes of machine learning, this research aims to mitigate risks associated with opaque algorithms, fostering greater security, user trust, and reliability in digital threat detection and prevention.