Skip to main content

Research Repository

See what's under the surface

Probabilistic word association for dialogue act classification with recurrent neural networks

Duran, Nathan; Battle, Steve

Authors

Nathan Duran Nathan.Duran@uwe.ac.uk
Graduate Tutor in Artificial Intelligence



Contributors

Giacomo Boracchi
Editor

Lazaros Iliadis
Editor

Chrisina Jayne
Editor

Aristidis Likas
Editor

Abstract

The identification of Dialogue Acts (DA) is an important aspect in determining the meaning of an utterance for many applications that require natural language understanding, and recent work using recurrent neural networks (RNN) has shown promising results when applied to the DA classification problem. This work presents a novel probabilistic method of utterance representation and describes a RNN sentence model for out-of-context DA Classification.The utterance representations are generated from keywords selected for their frequency association with certain DAs. The proposed probabilistic representations are applied to the Switchboard DA corpus and performance is compared with pre-trained word embeddings using the same baseline RNN model. The results indicate that the probabilistic method achieves 75.48% overall accuracy and an improvement over the word embedding representations of 1.8%. This demonstrates the potential utility of using statistical utterance representations, that are able to capture word-DA relationships,
for the purpose of DA classification.

Publication Date Apr 20, 2018
Peer Reviewed Peer Reviewed
Pages 229-239
Series Title Communications in Computer and Information Science
Book Title Engineering Applications of Neural Networks
ISBN 9783319651729
APA6 Citation Duran, N., & Battle, S. (2018). Probabilistic word association for dialogue act classification with recurrent neural networks. In C. Jayne, A. Likas, L. Iliadis, & G. Boracchi (Eds.), Engineering Applications of Neural Networks, 229-239. Springer
Publisher URL http://dx.doi.org/10.1007/978-3-319-65172-9
Additional Information Title of Conference or Conference Proceedings : Engineering Applications of Neural Networks

Files





You might also like



Downloadable Citations