What is it about?
Recommender systems rank among the top for driving traffic on YouTube, but how does it drive human attention toward videos? This paper presents a set of measurements that connect the structure of YouTube recommendation network to video popularity. We find a central cluster that occupies most of the attention is made out of videos that are mainly recommended among themselves. We also set up a task of estimating the influence between videos. Altogether, this paper provides a new set of tools to better understand the impacts of recommender systems on collective social attention.
Featured Image
Photo by NordWood Themes on Unsplash
Why is it important?
Understanding the effects of recommender systems is an important topic. Apart from the direct benefits to content owners and hosting sites, it helps the general users understand how their attention are shaped by the algorithmic recommendation, which in turn helps them be conscious of the relevance, novelty and diversity trade-offs in the content they are recommended to.
Perspectives
This paper is the result of a year-long project during my Ph.D grind. There are lots of hard work for creating the needed dataset for this study, especially when the concept of recommendation network is still relatively new. I hope this work can bring positive impacts to the community and inspire people to follow up the work in measuring the effects of recommender systems.
Siqi Wu
Australian National University
Read the Original
This page is a summary of: Estimating Attention Flow in Online Video Networks, Proceedings of the ACM on Human-Computer Interaction, November 2019, ACM (Association for Computing Machinery),
DOI: 10.1145/3359285.
You can read the full text:
Resources
Contributors
The following have contributed to this page
