Investigating the applications of artificial intelligence in cyber security

Naveed Naeem Abbas,Tanveer Ahmed,Syed Habib Ullah Shah,Muhammad Omar,Han Woo Park
DOI: https://doi.org/10.1007/s11192-019-03222-9
IF: 3.801
2019-09-09
Scientometrics
Abstract:Artificial Intelligence (AI) provides instant insights to pierce through the noise of thousands of daily security alerts. The recent literature focuses on AI's application to cyber security but lacks visual analysis of AI applications. Structural changes have been observed in cyber security since the emergence of AI. This study promotes the development of theory about AI in cyber security, helps researchers establish research directions, and provides a reference that enterprises and governments can use to plan AI applications in the cyber security industry. Many countries, institutions and authors are densely connected through collaboration and citation networks. Artificial neural networks, an AI technique, gave birth to today's research on cloud cyber security. Many research hotspots such as those on face recognition and deep neural networks for speech recognition may create future hotspots on emerging technology, such as on artificial intelligence systems for security. This study visualizes the structural changes, hotspots and emerging trends in AI studies. Five evaluation factors are used to judge the hotspots and trends of this domain and a heat map is used to identify the areas of the world that are generating research on AI applications in cyber security. This study is the first to provide an overall perspective of hotspots and trends in the research on AI in the cyber security domain.
information science & library science,computer science, interdisciplinary applications
What problem does this paper attempt to address?
The paper attempts to address the following issues: 1. **Structural Changes**: Investigating the structural changes in the field of cybersecurity research since the application of artificial intelligence (AI). These changes include shifts in research hotspots, directions, and trends. 2. **Impact of Collaboration and Citation Networks**: Exploring the significance and impact of tight and loose connections among journals, authors, institutions, and countries in collaboration and citation networks. By analyzing these networks, the interaction and collaboration patterns among different research entities can be understood. 3. **Application of Scientometric Visualization**: Proposing how to use scientometric visualization techniques to detect emerging trends of AI in the field of cybersecurity and generate research agendas. This includes methods such as co-citation reference networks, burst references, co-occurrence keyword networks, burst keywords, and dual-map overlay networks. Specifically, the study addresses these issues through the following methods: - **Data Collection**: Retrieving relevant literature from the Web of Science (WoS) database to ensure coverage of as many research outcomes as possible. - **Scientometric Analysis**: Using CiteSpace V software to generate co-citation reference networks, co-occurrence keyword networks, and journal dual-map overlay networks to identify structural changes and hotspots. - **Hotspot and Trend Analysis**: Revealing research hotspots and emerging trends through the analysis of the aforementioned networks, providing references for future research. The main purpose of this study is to provide references for theoretical research and practical applications, helping scholars determine future research directions of AI in the field of cybersecurity.