The US Twitter landscape isn’t as political as it’s made out to be

Written by Subhayan Mukerjee, Kokil Jaidka, Yphtach Lelkes for Nicolas Berube, La Presse CanadaBased on the findings of our paper, recently accepted in Political Communication Image: LaPresse Canada Were you surprised when you realized that politics isn't the main driver of the US Twitterverse?It’s somewhat reassuring to see that people on Twitter aren’t that different...

Thoughts on Vine vs TikTok

Compiled for Romano Santos, Vice. Image: CC Why do you think Vine was so popular, and why do you think it ultimately failed?  Vine demonstrated the power of unedited raw video in a social media space saturated with text and touched-up selfies. It created a new genre of fast comedy content that is still nostalgically...

The language of regional well-being

Jaidka, K., Giorgi, S., Schwartz, H. A., Kern, M. L., Ungar, L. H., & Eichstaedt, J. C. (2020). Estimating geographic subjective well-being from twitter: a comparison of dictionary and data-driven language methods. Proceedings of the National Academy of Sciences. In the COVID-19 era that we are living in today, social media posts can help us...

The #StayHomeforSG dashboard launch

Social media in COVID-19: building resilience or sowing stress? When the circuit-breaker came into effect in April 2020, Singaporeans suspended their daily lives to check the spread of the COVID-19 virus. Social media and computer-mediated-communication became the lifeline for communication as well as breaking news. However, social media was considered by many to be a...

The SMOL Project

Image credits: Swati Vats for the SMOL project. The Social Media, Online behavior and Language project focuses on online human behavior and computer mediated communication. SMOL is also internet lingo for something that's small and cute. Founding member and PI: Dr. Kokil Jaidka, Assistant Professor, NUS Research Areas: The role of affordances in computer-mediated communication:...

Language modeling on social media: the pros and cons

A number of studies have used social media language to (a) profile individuals, (b) profile linguistic styles, (c) profile communities, and (d) extrapolate the results to other domains, individuals, and communities. However, my work shows that pre-trained models may not scale well to other platforms, or even on the same platform to measure aggregated groups of...