Unsupervised Dialogue Act Induction using Gaussian Mixtures

Tomáš Brychcín, Pavel Král


Abstract
This paper introduces a new unsupervised approach for dialogue act induction. Given the sequence of dialogue utterances, the task is to assign them the labels representing their function in the dialogue. Utterances are represented as real-valued vectors encoding their meaning. We model the dialogue as Hidden Markov model with emission probabilities estimated by Gaussian mixtures. We use Gibbs sampling for posterior inference. We present the results on the standard Switchboard-DAMSL corpus. Our algorithm achieves promising results compared with strong supervised baselines and outperforms other unsupervised algorithms.
Anthology ID:
E17-2078
Volume:
Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers
Month:
April
Year:
2017
Address:
Valencia, Spain
Venue:
EACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
485–490
Language:
URL:
https://aclanthology.org/E17-2078
DOI:
Bibkey:
Cite (ACL):
Tomáš Brychcín and Pavel Král. 2017. Unsupervised Dialogue Act Induction using Gaussian Mixtures. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers, pages 485–490, Valencia, Spain. Association for Computational Linguistics.
Cite (Informal):
Unsupervised Dialogue Act Induction using Gaussian Mixtures (Brychcín & Král, EACL 2017)
Copy Citation:
PDF:
https://preview.aclanthology.org/update-css-js/E17-2078.pdf