Reproduce “More than Meets the Tie: Examining the Role of Interpersonal Relationships in Social Networks” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Minje Choi, Ceren Budak, Daniel M. Romero, David Jurgens
A study from the University of Michigan infers specific interpersonal relationship types from dyadic Twitter interactions, developing a neural network model that achieves a 0.70 macro F1 score in classifying five relationship categories. The research demonstrates that integrating these relationship types enhances retweet prediction, leading to a 2% increase in overall recall.
Key insights
- Different interpersonal relationship types on Twitter exhibit distinct and predictable patterns across linguistic, topical, network, and temporal dimensions, validating existing sociological theories.
- Explicit relationship types can be accurately inferred from online interaction data, enabling large-scale, relationship-aware network analyses.
- Understanding the specific nature of interpersonal ties is crucial for modeling complex social phenomena like information diffusion, as it influences what content is shared and how.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.