A group of students from DIS Copenhagen Spring 2019 were published in Medium, an online social journalism platform, for the work they did on their final project in the Computational Analysis of Big Data elective course. Students in the group included: Alex Sahai, Pomona College, Chadd Yuan, Tufts University, Lilian Gjertsson, Tufts University, Marissa Baker, Middlebury College, and Yongwook Nam, Amherst College.
The article is titled “Political Polarization on Twitter” and discusses the language that Trump uses on social media in comparison to other politicians. They also analyzed the meaning behind sentiment phrases like ‘crooked Hillary’ or subjective phrases like ‘fake news.’
‘What can be gleaned about President Trump from his midnight Twitter rampages? How about from a live stream of all mentions of Trump? And more generally, we wanted to get a landscape of Twitter as a global means for discussion…’ (Sahai, Yuan, Gjertsson, Baker, and Nam).
The students used various algorithms to track Trump’s language over the years and give the reader visual diagrams such as word clouds and sklearn’s CountVectorizer that present their findings.
For the conclusion of their project, the students compared Trump’s Twitter habits to that of DIS faculty member and Danish data scientist, Ulf Aslak Jensen. They explored the difference in sentiment and subjectivity used in Jensen’s tweets versus Trump’s. Turns out that Jensen, with his current social media presence of just 1,000 Twitter followers, probably wouldn’t win an election in the US any time soon.
Congratulations to Alex, Chadd, Lilian, Marissa, and Yongwook for their fascinating work.
Get More Information
Graphic provided by Alex Sahai, Chadd Yuan, Lilian Gjertsson, Marissa Baker, and Yongwook Nam.Back to all news