Leveraging SNS Data for E-Sports Recommendation: Analyzing Popularity and User Satisfaction Metrics †

Abstract

The rapid rise of social media and widespread Internet access have contributed significantly to the global popularity of e-sports. However, while popular e-sports attract considerable attention, niche e-sports remain underexplored, limiting user discovery and engagement. This paper proposes a Twitter-based recommendation system that uses advanced data management and processing techniques to address the challenge of identifying and recommending both popular and niche e-sports. The system analyzes social media metadata, including user IDs, followers, followees, engagements, and impressions, to calculate two critical metrics: popularity and satisfaction. Based on the combination of these metrics, the system calculates overall scores for each e-sports and generates two distinct rankings: one for popular and another for niche e-sports. The proposed system reflects the application of data-driven methodologies and social network analysis in creating recommendations that meet diverse user preferences, highlighting the relevance of data processing technologies in personalized content delivery. Experimental evaluations, using a dataset derived from Twitter hashtags (#) representing 30 target e-sports in 2022, demonstrate the system’s effectiveness in capturing the emerging dynamics in e-sports and providing actionable insights for diverse user preferences. This study highlights the potential of SNS-based technologies to advance data processing, analysis, and application within the e-sports ecosystem.