STEntConv: Predicting Disagreement with Stance Detection and a Signed Graph Convolutional Network

arXiv:2403.15885v1 Announce Type: new
Abstract: The rise of social media platforms has led to an increase in polarised online discussions, especially on political and socio-cultural topics such as elections and climate change. We propose a simple and novel unsupervised method to predict whether the authors of two posts agree or disagree, leveraging user stances about named entities obtained from their posts. We present STEntConv, a model which builds a graph of users and named entities weighted by stance and trains …

abstract arxiv authors change climate climate change cs.cl detection discussions elections graph media network novel platforms political simple social social media social media platforms topics type unsupervised