Analyzing Social Media Engagement of Computer Science Conferences

2025-03-02

AI Summary

This study investigates disparities in social media engagement across 22 computer science conferences on X (formerly Twitter), focusing on dissemination activity, user interaction, and content characteristics across subfields. Method: We employ large-scale data collection, statistical hypothesis testing (e.g., chi-square tests, ANOVA), and human annotation to conduct the first empirical, cross-conference comparison of content categories, sentiment polarity, and tweet length. Contribution/Results: Significant inter-conference differences emerge across all three dimensions (p < 0.01). “Likes” dominate user engagement—substantially outnumbering retweets and replies. Sentiment orientation exhibits strong domain specificity, correlating systematically with content type. The findings provide a reproducible, evidence-based framework for benchmarking and optimizing digital dissemination strategies across academic conferences, enabling cross-domain comparative analysis in scholarly communication research.

Technology Category

• Application Domains: Humanities & Computational Social Science
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Natural Language Processing: Sentiment Analysis, Stylistic Analysis, and Argument Mining

Application Category

• Web Mining and Content Analysis: Sentiment analysis and opinion mining
• Social Networks and Social Media: Computational social science
• Security and Privacy: Large-scale security measurements

Abstract

Context: X, formerly known as Twitter, is one of the largest social media platforms and has been widely used for communication during research conferences. While previous studies have examined how users engage with X during these events, limited research has focused on analyzing the content posted by computer science conferences. Objective: This study investigates how conferences from different areas of computer science perform on social media by analyzing their activity, follower engagement, and the content posted on X. Method: We collect posts from 22 computer science conferences and conduct statistical experiments to identify variations in content. Additionally, we perform a manual analysis of the top five posts for each engagement metric. Results: Our findings indicate statistically significant differences in category, sentiment, and post length across computer science conference posts. Among all engagement metrics, likes were the most common way users interacted with conference content. Conclusion: This study provides insights into the social media presence of computer science conferences, highlighting key differences in content, sentiment, and engagement patterns across different venues.

Problem

• Analyze social media engagement of computer science conferences.
• Examine content, sentiment, and engagement patterns on X.
• Identify variations in activity and follower engagement metrics.

Innovation

• Analyzed social media posts from 22 conferences
• Conducted statistical experiments on engagement metrics
• Manually analyzed top posts for engagement insights