Scale-Free Leadership Structure in Social Recommender Systems
Author Information
Author(s): Zhou Tao, Medo Matúš, Cimini Giulio, Zhang Zi-Ke, Zhang Yi-Cheng
Primary Institution: University of Electronic Science and Technology of China
Hypothesis
Can social recommender systems exhibit a scale-free leadership structure?
Conclusion
The study found that social recommender systems can develop a scale-free leadership structure where users with broad interests and good judgments become popular leaders.
Supporting Evidence
- The study analyzed networks from four leading social sites.
- It proposed an adaptive network model driven by social recommending.
- Simulations indicated that the recommendation mechanism improves user experience.
Takeaway
This study shows that in social networks, some users become popular leaders because they have a wide range of interests and make good recommendations.
Methodology
The study used empirical analysis of data from social bookmarking sites and agent-based simulations to model leadership structures.
Limitations
The model does not account for social factors like friendship and reciprocity.
Participant Demographics
Users from four social bookmarking sites: Delicious, Flickr, Twitter, and YouTube.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website