Invited Paper: Discourse Structures and Language Technologies

I want to tell a story about computational approaches to discourse structure. Like all such stories, it takes some liberty with actual events and times, but I think stories put things into perspective, and make it easier to understand where we are and how we might progress. Part 1 of the story (Section 2) is the past. Here we see early computational work on discourse structure aiming to assign a simple tree structure to a discourse. At issue was what its internal nodes corresponded to. The debate was fierce, and suggestions that other structures might be more appropriate were ignored or subjected to ridicule. The main uses of discourse structure were text generation and summarization, but mostly in small-scale experiments. Part 2 of the story (Section 3) is the present. We now see different types of discourse structure being recognized, though perhaps not always clearly distinguished. An increasing number of credible efforts are aimed at recognizing these structures automatically, though performance on unrestricted text still resembles that of the early days of robust parsing. Generic applications are also beginning to appear, as researchers recognize the value of these structures to tasks of interest to them. Part 3 of the story (Section 4) is the future. We now see the need for a mid-line between approaches hostage to theory and empirical approaches free of theory. An empirical approach underpinned by theory will not only motivate sensible back-off strategies in the face of unseen data, but also enable us to understand how the different discourse structures inter-relate and thereby to exploit their mutual recognition. This should allow more challenging applications, such as improving the performance of statistical machine translation (SMT) through the extended locality of discourse structures and the linguistic phenomena they correlate with.