A Model to Support Collective Reasoning: Formalization, Analysis and Computational Assessment

In this paper we propose a new model to represent human debates and methods obtain collective conclusions from them. This overcomes two drawbacks of existing approaches. First, our does not assume that participants agree on the structure debate. It by allowing express their opinion about all aspects Second, participants’ opinions are rational, an assumption significantly limits current Instead, define weaker notion rationality characterises coherent opinions, consider different scenarios based coherence individual level consensus. We provide formal analysis aggregation functions compute decision debate structure. particular, demonstrate aggregated can be even if there is lack consensus coherent. conclude with empirical evaluation demonstrating computed efficiently for real-sized debates.