Extracting group decision making experts' preferences from debate transcriptions using sentiment analysis

Abstract

Traditional group decision making methods work under the assumption that experts must provide their preferences to a computational system after carrying out a thorough debate. Nevertheless, this is not natural since the mean where experts express their ideas and preferences is inside the debate. In this paper, a novel method that is capable of extracting experts' preferences directly from the debate transcripts is presented. Thanks to this, experts do not have to go through a stressful process in order to provide the same information that they have already provide in the discussion part.