2017 #SMSociety Theme: Social Media for Social Good or Evil

Our online behaviour is far from virtual–it extends our offline lives. Much social media research has identified the positive opportunities of using social media; for example, how people use social media to form support groups online, participate in political uprising, raise money for charities, extend teaching and learning outside the classroom, etc. However, mirroring offline experiences, we have also seen social media being used to spread propaganda and misinformation, recruit terrorists, live stream criminal activities, reinforce echo chambers by politicians, and perpetuate hate and oppression (such as racist, sexist, homophobic, and anti-Semitic behaviour).

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

Session 4A [clear filter]
Sunday, July 30


Session 4A: Opinion Mining
avatar for Christoph Lutz

Christoph Lutz

Postdoc / Assistant Professor, BI Norwegian Business School
I am a researcher at BI Norwegian Business School Oslo and at the University of Leipzig. My work is in the field of new communication technologies and social media, where I investigate digital inequalities, online participation, trust, privacy and the sharing economy. I have a background... Read More →

Sunday July 30, 2017 11:00 - 12:30
TRS 1-073 - 7th Flr Ted Rogers School of Management, Ryerson University 55 Dundas Street West, Toronto, ON M5G 2C4


#DistractinglySexy: How Social Media Was Used As A Counter Narrative On Gender In STEM [FULL]
Authors: Ann Pegoraro, Emily Tetzlaff, Emily Jago and Tammy Eger

Abstract: On June 8th, 2015, Nobel laureate Sir Tim Hunt freely expressed his opinion on mixed-gender labs, while attending a President's lunch at the World Conference of Science Journalists:
“Let me tell you about my trouble with girls. Three things happen when they are in the lab: You fall in love with them, they fall in love with you, and when you criticize them they cry.”
In the days following his statement, the hashtag #DistractinglySexy trended on Twitter. The purpose of this study was to investigate how Twitter users interpreted the Tim Hunt speech, and how they represented their message through visual media on Twitter. The software program Hashtracking was used to gather 58,969 tweets that contained an image from the #DistractinglySexy hashtag. Content analysis was used to analyze the images collected and a codebook was developed through an adaptation of the ‘Draw-a-Scientist Test’ (DAST), a test initially designed to reveal children’s attitudes and beliefs about science through the use of stereotypical features. To enable human coding of such a large data set, a purposeful sample of 3,648 images was extracted for analysis. Intercoder reliability scores ranged from 0.84 to 1.0, all within the acceptable range. The results of this study indicated that users of the hashtag predominately portrayed themselves posed in personal protective equipment, in a laboratory setting. This study contributes to social media literature, by illustrating how this medium was utilized to create counter narratives that combat and highlight the challenges women in STEM face.

Sunday July 30, 2017 11:01 - 12:30
TRS 1-073 - 7th Flr Ted Rogers School of Management, Ryerson University 55 Dundas Street West, Toronto, ON M5G 2C4


Labels And Sentiment In Social Media: On The Role Of Perceived Agency In Online Discussions Of The Refugee Crisis [FULL]
Authors: Ju-Sung Lee and Adina Nerghes

Abtract: Focusing on the recent events in the Middle East, that have pushed many to flee their countries and seek refuge in neighboring countries or in Europe, we investigate dynamics of label use in social media, the emergent patterns of labeling that can cause further disaffection and tension, and the sentiments associated with the different labels. For this, we examine key labels pertaining to the refugee/migrant crisis and their usage in the user comment thread of a highly viewed and informational video of the crisis on YouTube. The use of labels indicate that migration issues are being framed not only through labels characterizing the crisis but also by their describing individuals themselves. The sentiments associated with these labels depart from what one would normally expect; in particular, negative sentiment is attached to labels that would otherwise be deemed neutral or positive. Interestingly, both positive and negative labels exhibit increased negativity across time. Using topic modeling and sentiment analysis jointly, we discover that the latent topics of the most positive comments show more overlap than those topics of the most negative comments, which are more focused and partitioned. In terms of sentiment, we find that labels indicating some degree of perceived agency or opportunity, such as 'migrant' or 'immigrant', are embedded in less sympathetic comments than those labels indicating a need to escape war-torn regions or persecution (e.g., asylum seeker or refugee). Our study offers valuable insights into the direction of public sentiment and the nature of discussions surrounding this significant societal event as well as the nature of online opinion sharing.

Sunday July 30, 2017 11:01 - 12:30
TRS 1-073 - 7th Flr Ted Rogers School of Management, Ryerson University 55 Dundas Street West, Toronto, ON M5G 2C4


Stance Classification Of Twitter Debates: The Encryption Debate As A Use Case [FULL]
Authors: Aseel Addawood, Jodi Schneider and Masooda Bashir

Abstract: Social media have enabled a revolution in user-generated content. They allow users to connect, build community, produce and share content, and publish opinions. To better understand online users’ attitudes and opinions, we use stance classification. Stance classification is a relatively new and challenging approach to deepen opinion mining by classifying a user's stance in a debate. Our stance classification use case is tweets that were related to the spring 2016 debate over the FBI’s request that Apple decrypt a user’s iPhone. In this “encryption debate,” public opinion was polarized between advocates for individual privacy and advocates for national security. We propose a machine learning approach to classify stance in the debate, and a topic classification that uses lexical, syntactic, Twitter-specific, and argumentative features as a predictor for classifications. Models trained on these feature sets showed significant increases in accuracy relative to the baseline.

Sunday July 30, 2017 11:01 - 12:30
TRS 1-073 - 7th Flr Ted Rogers School of Management, Ryerson University 55 Dundas Street West, Toronto, ON M5G 2C4