Enhancing Cybersecurity Knowledge Management through AI-Enabled Technologies
Adel Ismail Al‐Alawi, Mariam Al-Rashidi · 2024
This study explores the potential of knowledge management (KM) models in the field of artificial intelligence (AI)-powered cybersecurity, specifically in relation to machine learning, deep learning, and natural language processing approaches. This study systematically reviews literature since 2019 on algorithms in KM. Inclusion criteria reflect direct significance to algorithm-KM intersections, whereas exclusion criteria filter out pre-2019 studies and irrelevant papers. Using PubMed, IEEE Xplore, ScienceDirect, and Google Scholar, the search employed Boolean operators to refine queries with keywords such as "cybersecurity," "machine learning," "artificial intelligence," "knowledge management," and "algorithm."The paper evaluates various models, including convolutional neural networks (CNN), quasi-recurrent neural networks (QRNN), Principal Component Analysis (PCA), and Random Forest (RF), through an extensive analysis of literature in diverse areas. The study finds that CNN and QRNN perform better in tasks related to pattern recognition, PCA is useful in detecting anomalies, and RF is adaptable across a wide range of applications. The research provides valuable insights for companies seeking effective knowledge management methods, emphasizing the importance of flexibility and interpretability. Although individual models have limitations, this study highlights the mutually beneficial relationship between AI and KM, and how their collaboration can improve the overall cybersecurity situation.