Social Media Data Analysis. In Data Science & Engineering Master of Advanced Study (DSE MAS) Capstone Projects

This project focuses on studying how Twitter images impact the narrative of hashtags. A lot of research in Twitter data has been focused on separate textual content without media attachments. Images attached to a tweet provide an additional dimension to help understand a tweet’s context and the user’s general opinion. The assumption is that the visual images reinforce the opinion that was presented in the text. The project aims at finding specific patterns in tweets where media files (images) are used to change the narrative of the corresponding hashtag and co-occurred hashtags. This is achieved by studying the topic of solo hashtag and co-occurred hashtags without or with associated media files. Media file as a visual channel is a powerful medium and can change the subject and original purpose of a hashtag for a given audience. A small number of media tweets (images) associated with a hashtag can have a higher influential impact on an observer/user than the same or even higher number of tweets without media. There are multiple benefactors to the findings from this project. 1- Election organizer as part of a campaign can study and detect endorsing and opposing trends and act by counter measures using similar techniques. 2- Social Media platform and specially Twitter itself can detect patterns and potentially restrict the behavior. 3- Journalists can report to general public on how a potential small group of influencers can sway a narrative and push various agendas.