Time-based Gap Analysis of Cybersecurity Trends in Academic and Digital Media

Abstract

This study analyzes cybersecurity trends and proposes a conceptual framework to identify cybersecurity topics of social interest and emerging topics that need to be addressed by researchers in the field. The insights drawn from this framework allow for a more proactive approach to identifying cybersecurity patterns and emerging threats that will ultimately improve the collective cybersecurity posture of the modern society. To achieve this, cybersecurity-oriented content in both media and academic corpora, disseminated between 2008 and 2018, were morphologically analyzed via text mining. A total of 3,556 academic papers obtained from the top-10 highly reputable cybersecurity academic conferences, and 4,163 news articles collected from the New York Times were processed. The LDA topic modeling followed optimal perplexity and coherence scores resulted in 12 trendy topics. Next, the time-based gap between these trendy topics was analyzed to measure the correlation between media and trendy academic topics. Both convergences and divergences between the two cybersecurity corpora were identified, suggesting a strong time-based correlation between these resources. This framework demonstrates the effective use of automated techniques to provide insights about cybersecurity topics of social interest and emerging trends and informs the direction of future academic research in this field.