Topic switch models for dialogue management in virtual humans

This paper presents a novel data-driven Topic Switch Model based on a cognitive representation of a limited set of topics that are currently in-focus, which determines what utterances are chosen next. The transition model was statistically learned from a large set of transcribed dyadic interactions....

Full description

Bibliographic Details
Main Authors:	Zhu, Wenjue, Chowanda, Andry, Valstar, Michel F.
Format:	Conference or Workshop Item
Language:	English
Published:	2016
Online Access:	http://eprints.nottingham.ac.uk/35622/ http://eprints.nottingham.ac.uk/35622/ http://eprints.nottingham.ac.uk/35622/1/topic-switch-models.pdf

id	nottingham-35622
recordtype	eprints
spelling	nottingham-356222017-10-15T23:13:52Z http://eprints.nottingham.ac.uk/35622/ Topic switch models for dialogue management in virtual humans Zhu, Wenjue Chowanda, Andry Valstar, Michel F. This paper presents a novel data-driven Topic Switch Model based on a cognitive representation of a limited set of topics that are currently in-focus, which determines what utterances are chosen next. The transition model was statistically learned from a large set of transcribed dyadic interactions. Results show that using our proposed model results in interactions that on average last 2.17 times longer compared to the same system without our model. 2016-09-20 Conference or Workshop Item PeerReviewed application/pdf en http://eprints.nottingham.ac.uk/35622/1/topic-switch-models.pdf Zhu, Wenjue and Chowanda, Andry and Valstar, Michel F. (2016) Topic switch models for dialogue management in virtual humans. In: 16th International Conference on Intelligent Virtual Agents (IVA 2016), 20-23 Sept, 2016, Los Angeles, California, USA. http://www.springer.com/us/book/9783319476643
repository_type	Digital Repository
institution_category	Local University
institution	University of Nottingham Malaysia Campus
building	Nottingham Research Data Repository
collection	Online Access
language	English
description	This paper presents a novel data-driven Topic Switch Model based on a cognitive representation of a limited set of topics that are currently in-focus, which determines what utterances are chosen next. The transition model was statistically learned from a large set of transcribed dyadic interactions. Results show that using our proposed model results in interactions that on average last 2.17 times longer compared to the same system without our model.
format	Conference or Workshop Item
author	Zhu, Wenjue Chowanda, Andry Valstar, Michel F.
spellingShingle	Zhu, Wenjue Chowanda, Andry Valstar, Michel F. Topic switch models for dialogue management in virtual humans
author_facet	Zhu, Wenjue Chowanda, Andry Valstar, Michel F.
author_sort	Zhu, Wenjue
title	Topic switch models for dialogue management in virtual humans
title_short	Topic switch models for dialogue management in virtual humans
title_full	Topic switch models for dialogue management in virtual humans
title_fullStr	Topic switch models for dialogue management in virtual humans
title_full_unstemmed	Topic switch models for dialogue management in virtual humans
title_sort	topic switch models for dialogue management in virtual humans
publishDate	2016
url	http://eprints.nottingham.ac.uk/35622/ http://eprints.nottingham.ac.uk/35622/ http://eprints.nottingham.ac.uk/35622/1/topic-switch-models.pdf
first_indexed	2018-09-06T12:36:50Z
last_indexed	2018-09-06T12:36:50Z
_version_	1610861679526019072

Topic switch models for dialogue management in virtual humans

Similar Items