Multi-language IoT information security standard item matching based on deep learning

Abstract

In the realm of IoT information security and other domains, various information security standards exist, such as the IEC 62443 series standards published by the International Electrotechnical Commission and ISO/IEC 27001 by the International Organization for Standardization. Business organizations are striving to improve and protect their operations through the implementation and study of these information security standards. However, comparing or pinpointing applicable control measures is becoming increasingly labor-intensive and prone to errors or deviations, especially given the plethora of information standards available. Identifying specific control measures scattered across different information security standards is gradually becoming an important issue. In this research, we utilise a range of domestic and international information security standards as the foundation, employing text mining and deep learning methods to map the similar parts of control measures between standards, thereby enhancing the efficiency of comparison tasks and allowing human resources to be allocated to more pertinent issues.