Ng, Stephanie https://orcid.org/0009-0009-1416-7614
Zhang, James
Yu, Samson
Bhatti, Asim
Backholer, Kathryn
Lim, C. P.
Funding for this research was provided by:
Deakin University
Article History
Received: 30 July 2024
Accepted: 2 February 2025
First Online: 3 March 2025
Declarations
:
: On behalf of all authors, the corresponding author states that there is no conflict of interest.
: Our research is focused on political text scrutiny and discourse evaluation on publicly available data using computational approaches. While our devised approach could efficiently aid researchers to gain valuable insights in political text analysis, it is important to note potential biases inherent in both training data and model itself, which could lead to a bias evaluation of political discourse. One notable limitation of our approach is the unintended truncation of dialogues or utterances, which could result in potential loss of context and meaning of the original messages. We also acknowledge that these models could potentially influence public opinion and shape political narratives, which could result in unintended consequences such as the spread of misinformation when the results are misinterpreted. As such, it is necessary to conduct further research for addressing these ethical considerations and improving model transparency and interpretability.