Neural Networks Toward Cybersecurity: Domain Map Analysis of State-of-the-Art Challenges

The growing interest in applying neural networks for cybersecurity has prompted a substantial increase in related research. This paper presents a comprehensive bibliometric analysis of research on cybersecurity towards neural networks published in the Web of Science over the past two decades (2003–2023) using bibliometric methods and CiteSpace software. The analysis encompasses yearly publication trends, types of publications, and trends across various dimensions such as publishing sources, organizations, researchers, countries, and keywords. Additionally, timeline and burst detection analyses were conducted to identify significant topic trends and citations in the last two decades. It also outlines the latest trends, under-explored topics, and open challenges.