Publication Type

Conference Proceeding Article


Publisher’s Version

Publication Date



Social media has become a popular platform for people toshare opinions. Among the social media mining researchprojects that study user opinions and issues, most focus onanalyzing posted and shared content. They could run into thedanger of non-representative findings as the opinions of userswho do not post content are overlooked, which often happensin today’s marketing, recommendation, and social sensing research.For a more complete and representative profiling ofuser opinions on various topical issues, we need to investigatethe opinions of the users even when they stay silent onthese issues. We call these users the issue specific-silent users(i-silent users). To study them and their opinions, we conductan opinion survey on a set of users for two popular social mediaplatforms, Twitter and Facebook. We further analyze theircontributed personal social media data. Our main findings arethat more than half of our users who are interested in issuei are i-silent users in Twitter. The same has been observedfor our Facebook users. i-silent users are likely to have differentopinion distribution from the users who post about i.With the ground truth user opinions from the survey, we furtherdevelop and apply opinion prediction methods to i-silentusers in Twitter and Facebook using their social media dataand their opinions on issues other than i.


Databases and Information Systems | Social Media

Research Areas

Data Science and Engineering


Proceedings of the 10th International AAAI Conference on Web and Social Media ICWSM 2016: Cologne, Germany, May 17–20

First Page


Last Page






City or Country

Palo Alto, CA

Copyright Owner and License


Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Additional URL