Revisiting Information Diffusion Beyond Explicit Social Ties: A Study of Implicit-Link Diffusion on Twitter

2025-05-13

AI Summary

This paper addresses the underexplored “implicit links” problem in social media information diffusion—i.e., content exposure and re-sharing by users via non-follower channels such as search engines, algorithmic recommendations, or external websites. Leveraging four large-scale Twitter datasets, we systematically define, identify, and quantify implicit-link-driven diffusion through propagation path tracing, network topology analysis, and user behavior clustering. Our results show that although implicit links contribute modestly to overall cascade size, they dominate cross-community and long-range diffusion; exhibit strong homophilous channel preferences; and enable identification of two distinct user types—explicit-link-dominant and implicit-link-dominant—whose sociodemographic and behavioral attributes differ systematically. This work reveals a dual-path diffusion mechanism in the algorithmic era, fundamentally extending classical social diffusion theory beyond follower-based networks.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Application Domains: Social Networks
• Humans and AI: Crowd Sourcing and Human Computation

Application Category

• Web Mining and Content Analysis: Content-based information diffusion
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks
• User Modeling, Personalization and Recommendation: Social recommender systems and personalization for the social Web

Abstract

Information diffusion on social media platforms is often assumed to occur primarily through explicit social connections, such as follower or friend relationships. However, information frequently propagates beyond these observable ties -- via external websites, search engines, or algorithmic recommendations -- forming implicit links between users who are not directly connected. Despite their potential impact, the mechanisms and characteristics of such implicit-link diffusion remain underexplored. In this study, we investigate the dynamics of nontrivial information diffusion mediated by implicit links on Twitter, using four large-scale datasets. We define implicit-link diffusion as the reposting of content by users who are not explicitly connected to the original poster. Our analysis reveals that users located farther from the original source in the social network are more likely to engage in diffusion through implicit links, suggesting that such links often arise from sources outside direct social relationships. Moreover, while implicit links contribute less to the overall diffusion size than explicit links, they play a distinct role in disseminating content across diverse and topologically distant communities. We further identify user groups who predominantly engage in diffusion through either explicit or implicit links, and demonstrate that the choice of diffusion channel exhibits strong patterns of social homophily. These findings underscore the importance of incorporating implicit-link dynamics into models of information diffusion and social influence.

Problem

• Investigates implicit-link information diffusion on Twitter
• Examines mechanisms of diffusion beyond explicit social ties
• Analyzes role of implicit links in cross-community dissemination

Innovation

• Analyzing implicit-link diffusion on Twitter
• Using four large-scale datasets for study
• Identifying distinct roles in community dissemination