This article presents a systematic literature review examining the application of Natural Language Processing (NLP) techniques to information security risk classification and identification in textual documents. The review follows the PRISMA methodology, aiming to identify primary approaches, algorithms, and challenges in applying NLP for automated analysis of security-related documents. Results reveal an increasing trend towards Transformer-based models, particularly BERT, for risk classification, showing significant improvements in accuracy compared to traditional methods. This review contributes to the development of automated information security risk classification systems, providing an initial foundation for future research in this emerging specialized domain.
Ahmed et al. (Thu,) studied this question.