Stance-based Argument Mining – Modeling Implicit Argumentation Using Stance
Especially social media contains a huge amount of noisy and implicit argumentation. We argue, that a stance-based model of argumentation has several advantages for mining this argumentation. In principle, stance-based argument mining assumes, that in a debate there are actually several sub debates. By detecting explicitly expressed stances regarding the sub debates, it should be possible to infer a robust proxy for the arguments in a debate.
For details and experiments see the following paper:
The data can be accessed for scientific purposes: