Paper: Automatic Extraction of Cue Phrases for Cross-Corpus Dialogue Act Classification

ACL ID C10-2150
Title Automatic Extraction of Cue Phrases for Cross-Corpus Dialogue Act Classification
Venue International Conference on Computational Linguistics
Session Poster Session
Year 2010
Authors

In this paper, we present an investiga- tion into the use of cue phrases as a ba- sis for dialogue act classification. We de- fine what we mean by cue phrases, and de- scribe how we extract them from a manu- ally labelled corpus of dialogue. We de- scribe one method of evaluating the use- fulness of such cue phrases, by applying them directly as a classifier to unseen ut- terances. Once we have extracted cue phrases from one corpus, we determine if these phrases are general in nature, by applying them directly as a classification mechanism to a different corpus to that from which they were extracted. Finally, we experiment with increasingly restric- tive methods for selecting cue phrases, and demonstrate that there are a small number of core cue phrases that are use- ful for dialogue act c...