Extracting decisions from multi-party dialogue using directed graphical models and semantic similarity

Trung Bui, Matthew Frampton, John Dowding and Stanley Peters

SIGDIAL Workshop on Discourse and Dialogue (SIGDIAL 2009)
Queen Mary University of London, September 11-12, 2009


We investigate the automatic detection of decision discussions in multi-party dialogue using directed graphical models (DGMs). Our approach distinguishes between different dialogue act (DA) types based on their roles in the formulation of a decision. DGMs enable us to model sequence and dependencies flexibly. We summarize decisions by extracting suitable words/phrases from DAs that concern the issue under discussion and its resolution. We use a semantic-similarity metric here to improve results on both manual and ASR transcripts.