Algorithmic impacts on social media

Abstract

Although some argue that curation algorithms can help to deal with information overload and improve user experience on social media platforms, they could also cause undesirable societal outcomes. This dissertation tries to investigate two important societal issues that can be influenced by curation algorithms on social media: (1) whether the curation algorithms limit the information an individual user consumes on social media, creating the “filter bubble” phenomenon, and lead to heightened polarization; (2) whether the curation algorithms contributes further the attention inequality on social media. We leveraged our unique dataset to carry out a difference-in-difference analysis by comparing users’ behavioral changes on these two platforms. The first study provides strong evidence for the creation of filter bubble caused by curation algorithms – Users consumed significant less diverse content and less attitude-challenging viewpoints after the implementation of curation algorithms. Furthermore, filter bubbles significantly aggravated general attitude polarization on social media as well as attitude polarization over public issues, and reduced users’ contribution to the platform as the number of posts generated by users, both reposted and original, dropped significantly. The second study answers the question whether such algorithms would aggravate the attention inequality on social media, our findings suggest that the implementation of curation algorithm (1) suppresses fan attraction and interaction reception for unpopular users, (2) helps to attract more discussion for unpopular topics at the topic level.

Date of Award	2020
Original language	English
Awarding Institution	The Hong Kong University of Science and Technology