07/06/2020

Characterizing Collective Attention via Descriptor Context: A Case Study of Public Discussions of Crisis Events

Ian Stewart, Diyi Yang, Jacob Eisenstein

Keywords: attention, cases, contexts, discussions, engagement, events, facebook, facebook twitter, factors, languages, large_scale, linguistic, locations, news, spread, topic, twitter

Abstract: Collective attention, i.e. the public attention paid to a particular topic, is a key factor in understanding how emerging topics and breaking news spread in online discussions. In most research, collective attention on social media is measured via aggregate metrics, such as the number of posts that mention a given name. However, collective attention is expressed not only in frequency, but also in content: linguistic features of the description of events and names reflect how writers expect their readers to perceive the information being discussed. In this work, we conduct a large-scale language analysis of public online discussions of breaking news events on Facebook and Twitter, focusing on five recent crisis events. We examine how people refer to locations, focusing specifically on contextual descriptors, such as ``San Juan´´ versus ``San Juan, Puerto Rico.´´ We find that the use of such descriptors is associated with proxies for social and informational expectations, including macro-level factors like the location´s global importance and micro-level factors like audience engagement. We also find a consistent decrease in descriptor context use over time at a collective level, particularly for less active authors. These insights provide evidence for theories about information expectations in public discussions, and they inform how researchers and crisis response organizations can better understand public perception of crisis events as they unfold.

 0
 0
 0
 0
This is an embedded video. Talk and the respective paper are published at ICWSM 2020 virtual conference. If you are one of the authors of the paper and want to manage your upload, see the question "My papertalk has been externally embedded..." in the FAQ section.

Comments

Post Comment
no comments yet
code of conduct: tbd Characters remaining: 140

Similar Papers