Simultaneous Dialogue Act segmentation and labelling using lexical and syntactical features

Ramon Granell, Stephen Pulman and Carlos D. Martinez-Hinarejos

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


Segmentation of utterances and annotation as Dialogue Acts can be helpful for several modules of Dialogue Systems. In this work, we study a statistical machine learning model to perform simultaneously these tasks using lexical features and incorporating deterministic syntactic restrictions. There is a slight improvement in both segmentation and labelling due to these restrictions.