Abstract

Online forums are commonplace today and used for various purposes: product support and troubleshooting, opining about events and people, and student interaction on online course platforms. Threads in these forums become long, involve posts from multiple users, and the chronological order of the posts in a thread does not represent a continuous flow of dialog. Adding structure to these threads is important for tasks such as information extraction, search, and summarization. One such aspect of structure is topic: a discussion often involves a number of subtopics. In this talk, I will present work aiming to characterize the content of a forum thread as a conversation tree of topics.

Our models jointly perform two tasks: (i) segment a thread into sub-parts, and (ii) assign a topic to each part. The key idea in this work is a formalization of topic structure using probabilistic grammars. By leveraging the flexibility of two grammar formalisms, Context-Free Grammars and Linear Context-Free Rewriting Systems, our models create desirable structures for forum threads: our topic segmentation is hierarchical, links non-adjacent segments on the same topic, and jointly labels the topic during segmentation. I will show that these models outperform a number of tree generation baselines.