top of page

Project Showcase

This was exploratory and experimental research to find out which topics stood out in Ugandan media coverage of Covid-19 according to a machine learning method called LDA Topic Modelling, which employs a Natural Language Processing (NLP) technique.

This study extends the organizational-centered transnational advocacy network by presenting how grassroots users strategically utilize social media platforms for achieving their diplomatic engagement with foreign actors. The network analysis and natural language processing of Twitter outreach on Hong Kong protests (N = 88,800) identify the key opinion leaders and the grassroots narratives under three core themes: geopolitics, universal values and humanitarian concern. The low threshold of Twitter participation provides extra direct channels for grassroots users to engage with foreign politicians and get their narratives heard.

This study uses data from TikTok (N = 8,173) to examine how short-form video platforms challenge the protest paradigm established by the mainstream media in the Black Lives Matter movement, which was triggered by George Floyd's death on 25 May 2020. A computer-mediated visual analysis is employed to identify the presence of four visual frames of protest (riot, confrontation, spectacle, and debate) in multimedia content.

This study selected popular conspiracies reflecting different political ideologies and conducts multiple survey rounds (N=500), comparing and contrasting the effect of different partisan affiliations on conspiracy endorsement.

bottom of page