PSentScore: Evaluating Sentiment Polarity in Dialogue Summarization

arXiv:2307.12371v2 Announce Type: replace
Abstract: Automatic dialogue summarization is a well-established task with the goal of distilling the most crucial information from human conversations into concise textual summaries. However, most existing research has predominantly focused on summarizing factual information, neglecting the affective content, which can hold valuable insights for analyzing, monitoring, or facilitating human interactions. In this paper, we introduce and assess a set of measures PSentScore, aimed at quantifying the preservation of affective content in dialogue summaries. Our findings …

abstract arxiv conversations cs.cl dialogue however human information insights monitoring research sentiment summarization summarizing textual type