Integrating natural language processing and pragmatic argumentation theories for argumentation support

摘要

Natural language processing (NLP) research and design that aims to model and detect opposition in text for the purpose of opinion classification, sentiment analysis, and meeting tracking, generally excludes the interactional, pragmatic aspects of online text. We propose that a promising direction for NLP is to incorporate the insights of pragmatic, dialectical theories of argumentation to more fully exploit the potential of NLP to offer sound, robust systems for various kinds of argumentation support.