A semantic framework for modelling quotes in email conversations

Abstract

Quoting is a common practice in online conversations, notably in email discussions and bulletin boards. Despite many applications that offer special functionalities to handle them, there is no agreement on how to semantically represent these quotes, so that they could be queried and analysed uniformly whatever the original application is. While various Semantic Web models, notably SIOC, have become popular for interlinking, exporting and exchanging information about online communities and their conversations, they do not provide means to model such quotes. In this paper, we present (i) an OWL2 extension of SIOC for representing quotes and (ii) a framework for automated quotes extraction from email archives. In addition, we demonstrate the utility of such formal representation for querying email archives with finergrained queries and for mining social networks based on the argumentative structure of email conversations.