Discourse Processing and Its Applications --- Tutoral at ACL-2019

Time: 28 July 9:00 - 12:30
Location: TBA

Slides: Part-1, Part-2

Tutors
Shafiq Joty Giuseppe Carenini Raymond T Ng Gabriel Murray

Tutorial Abstract

Discourse processing is a suite of Natural Language Processing (NLP) tasks to uncover linguistic structures from texts at several levels, which can support many downstream applications. This involves identifying the topic structure, the coherence structure, the coreference structure, and the conversation structure for conversational discourse. Taken together, these structures can inform text summarization, machine translation, essay scoring, sentiment analysis, information extraction, question answering, and thread recovery. The tutorial starts with an overview of basic concepts in discourse analysis – monologue vs. conversation, synchronous vs. asynchronous conversation, and key linguistic structures in discourse analysis. We also give an overview of linguistic structures and corresponding discourse analysis tasks that discourse researchers are generally interested in, as well as key applications on which these discourse structures have an impact.

Tutorial Outline

Introduction [25 mins]

  • Discourse & its different forms
  • Two discourse phenomena
  • Linguistic structures in discourse & discourse analysis tasks
  • Applications of discourse analysis

Coherence Structure, Corpora & Discourse Parsing [45 mins]

  • Discourse theories & coherence relations
  • Discourse parsing with RST
  • Discourse parsing in PDTB
  • Final remarks

Coffee Break [15 mins]

Coherence Models & Applications of Discourse [45 mins]

  • Overview of coherence models
  • Evaluation tasks
  • Applications of discourse

Conversational Structure [35 mins]

  • Discourse Structures in Conversations
  • Thread identification models for synchronous & asynchronous conversations
  • Speech act recognition models for synchronous & asynchronous conversations
  • Evaluation & Applications

Future Challenges [15 mins]

  • Learning from limited annotated data
  • Language & domain transfer
  • Discourse generation
  • New emerging applications