Ryan, Arthur (2024) Co-pilot widget for assisting the public in processing US presidential political candidate tweets from Twitter in 2024 US elections candidate choice through sarcasm and stance detection. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (746kB) | Preview |
Preview |
PDF (Configuration Manual)
Download (3MB) | Preview |
Abstract
With 50 percent of the world’s population going to the polls to elect new leaders in 2024 it was thought to be relevant to examine what can be done to lessen polarisation and improve quality of discourse online. The gap the paper intends to fill is the prior lack of combined natural language processing (NLP) techniques collected together to raise understanding of media. NLP results were 0.9960 for truth-fake detection, 0.9835 for sarcasm detection and 0.9978 for stance. Additionally, a set of two versions of graphic outcome to make understanding simpler and faster.
Actions (login required)
![]() |
View Item |