Abstract
The continuous growth of data volumes and the increasing complexity of modern IT infrastructures have significantly reduced the effectiveness of conventional approaches to auditing Information Security Management Systems. In such conditions, ensuring scalability and rapid analysis of large datasets becomes a critical challenge for security auditing processes. The purpose of this study is to examine the potential of applying machine learning (ML), deep learning (DL), and natural language processing (NLP) techniques to enhance and automate information security audit procedures. The study reviews contemporary algorithms used for security log analysis, anomaly detection, event correlation, and evaluation of organizational compliance with the requirements of the international standard ISO/IEC 27001. Based on the research, this study proposes an original conceptual model of intelligent auditing, built on a multi-layered architecture that integrates ML and Large Language Model (LLM) methods. The obtained results indicate that the integration of intelligent analytical techniques improves the accuracy of data analysis, supports automated identification of compliance gaps, and enables the transition from periodic assessments to continuous information security auditing.