Research and Projects
My area of research is Natural Language Processing (NLP), with a focus on analytical multi-party dialogue. In particular, I study civil, constructive dialogue with the aim of applying my findings to create better discussion platforms, dialogue agents, and tools. The path to improving these tools lies in better organizing these dialogues and understanding the context in which conversation is grounded.
Supreme Court Arguments
One of my projects involves Supreme Court data. If you have never sat through a full Supreme Court oral argument, you are missing out! Outside of enjoying the banter between Justices Breyer and Scalia, our aim is to understand how the Counsel present their arguments to the court and how Justices respond.
Criteria-based Writing Assessment
While dialogue is a key component of much of my work, the text in which some dialogue is grounded also has properties which interest me. In this second project, I work with Esin Durmuş on constructing a new framework for writing assessment.
Visual question answering has emerged from the intersection of computer vision and NLP. In this project we explore the balance between text and imagery.