journal article

Public sentiment and engagement on cybersecurity: Insights from Reddit discussions

Achuthan K.; Khobragade S.; Kowalski R.

Publication year: 2025
Source / journal: computers in human behavior reports
Country of origin: India, USA
Language: English
Authors’ affiliations: Center for Cybersecurity Systems and Networks, Amrita Vishwa Vidyapeetham, Amritapuri, Kerala, Kollam, 690525, India, College of Arts and Humanities, Clemson University, 105 Sikes Hall, Clemson, 29634, SC, United States
Import source: Scopus export

Abstract

Cyber insecurity is recognized amongst the top severe global risks by the World Economic Forum.This study addresses a significant gap in our understanding of the social dynamics of public perception and online discourse related to cybersecurity, with a particular focus on how social media platforms shape and reflect public engagement. We employ sentiment analysis, topic modeling, and zero-shot text classification to analyze 33,893 news and user posts, along with the corresponding 194,591 news and user comments, across five subreddit discussion forums between 2012 and 2022. Through the lens of the Social Amplification of Risk Framework (SARF) and Protection Motivation Theory (PMT), we uncover patterns in topic engagement, emotional responses, community support, and the interplay of sentiment and comment types. Our findings reveal a complex landscape of public engagement with cybersecurity news, characterized by high interest in government surveillance, cyber warfare, and software vulnerabilities. Comments vary across topics, with technical advice dominating discussions, while skepticism and concern are prevalent in privacy-related areas. Community reactions tend towards critical engagement, with comments expressing concern, disappointment, and anger receiving high upvotes. Post sentiment weakly correlates with comment types, where more negative posts elicit increased expressions of concern, anger, and skepticism. The community's dynamic engagement with cybersecurity issues underscores the need for tailored communication strategies that acknowledge both emotional responses and the desire for practical solutions. This study contributes to an understanding of public engagement with cybersecurity issues on social media while offering insights for stakeholders seeking to foster informed, resilient, and engaged online communities. © 2024 The Authors

Author keywords

artificial intelligence, bertopic modeling, cybercrime, public sentiment, reddit social media, text classification

Open DOI