A tractable DDN-POMDP Approach to Affective Dialogue Modeling for General Probabilistic Frame-based Dialogue Systems

B.H.T. Bui Huu Trung, Mannes Poel, Antinus Nijholt, Jakob Zwiers

Abstract

We propose a new approach to developing a tractable affective dialogue model for general probabilistic frame-based dialogue systems. The dialogue model, based on the Partially Observable Markov Decision Process (POMDP) and the Dynamic Decision Network (DDN) techniques, is composed of two main parts, the slot level dialogue manager and the global dialogue manager. Our implemented dialogue manager prototype can handle hundreds of slots; each slot might have many values. A first evaluation of the slot level dialogue manager (1-slot case) showed that with a 95% confidence level the DDN-POMDP dialogue strategy outperforms three simple handcrafted dialogue strategies when the user's action error is induced by stress.
Original languageUndefined
Title of host publicationProceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems
EditorsD. Traum, J. Alexandersson, A. Jonsson, I. Zukerman
Place of PublicationHyderabad, India
PublisherIJCAI
Pages34-37
Number of pages4
ISBN (Print)not assigned
StatePublished - 8 Jan 2007

Publication series

Name
PublisherIJCAI
Number1

Fingerprint

Managers

Keywords

  • HMI-MI: MULTIMODAL INTERACTIONS
  • HMI-SLT: Speech and Language Technology
  • METIS-241912
  • EWI-8523
  • IR-66715

Cite this

Bui Huu Trung, B. H. T., Poel, M., Nijholt, A., & Zwiers, J. (2007). A tractable DDN-POMDP Approach to Affective Dialogue Modeling for General Probabilistic Frame-based Dialogue Systems. In D. Traum, J. Alexandersson, A. Jonsson, & I. Zukerman (Eds.), Proceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems (pp. 34-37). Hyderabad, India: IJCAI.

Bui Huu Trung, B.H.T.; Poel, Mannes; Nijholt, Antinus; Zwiers, Jakob / A tractable DDN-POMDP Approach to Affective Dialogue Modeling for General Probabilistic Frame-based Dialogue Systems.

Proceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems. ed. / D. Traum; J. Alexandersson; A. Jonsson; I. Zukerman. Hyderabad, India : IJCAI, 2007. p. 34-37.

Research output: Scientific - peer-reviewConference contribution

@inbook{56046c32535b46d99f83b57f195a2087,
title = "A tractable DDN-POMDP Approach to Affective Dialogue Modeling for General Probabilistic Frame-based Dialogue Systems",
abstract = "We propose a new approach to developing a tractable affective dialogue model for general probabilistic frame-based dialogue systems. The dialogue model, based on the Partially Observable Markov Decision Process (POMDP) and the Dynamic Decision Network (DDN) techniques, is composed of two main parts, the slot level dialogue manager and the global dialogue manager. Our implemented dialogue manager prototype can handle hundreds of slots; each slot might have many values. A first evaluation of the slot level dialogue manager (1-slot case) showed that with a 95% confidence level the DDN-POMDP dialogue strategy outperforms three simple handcrafted dialogue strategies when the user's action error is induced by stress.",
keywords = "HMI-MI: MULTIMODAL INTERACTIONS, HMI-SLT: Speech and Language Technology, METIS-241912, EWI-8523, IR-66715",
author = "{Bui Huu Trung}, B.H.T. and Mannes Poel and Antinus Nijholt and Jakob Zwiers",
year = "2007",
month = "1",
isbn = "not assigned",
publisher = "IJCAI",
number = "1",
pages = "34--37",
editor = "D. Traum and J. Alexandersson and A. Jonsson and I. Zukerman",
booktitle = "Proceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems",

}

Bui Huu Trung, BHT, Poel, M, Nijholt, A & Zwiers, J 2007, A tractable DDN-POMDP Approach to Affective Dialogue Modeling for General Probabilistic Frame-based Dialogue Systems. in D Traum, J Alexandersson, A Jonsson & I Zukerman (eds), Proceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems. IJCAI, Hyderabad, India, pp. 34-37.

A tractable DDN-POMDP Approach to Affective Dialogue Modeling for General Probabilistic Frame-based Dialogue Systems. / Bui Huu Trung, B.H.T.; Poel, Mannes; Nijholt, Antinus; Zwiers, Jakob.

Proceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems. ed. / D. Traum; J. Alexandersson; A. Jonsson; I. Zukerman. Hyderabad, India : IJCAI, 2007. p. 34-37.

Research output: Scientific - peer-reviewConference contribution

TY - CHAP

T1 - A tractable DDN-POMDP Approach to Affective Dialogue Modeling for General Probabilistic Frame-based Dialogue Systems

AU - Bui Huu Trung,B.H.T.

AU - Poel,Mannes

AU - Nijholt,Antinus

AU - Zwiers,Jakob

PY - 2007/1/8

Y1 - 2007/1/8

N2 - We propose a new approach to developing a tractable affective dialogue model for general probabilistic frame-based dialogue systems. The dialogue model, based on the Partially Observable Markov Decision Process (POMDP) and the Dynamic Decision Network (DDN) techniques, is composed of two main parts, the slot level dialogue manager and the global dialogue manager. Our implemented dialogue manager prototype can handle hundreds of slots; each slot might have many values. A first evaluation of the slot level dialogue manager (1-slot case) showed that with a 95% confidence level the DDN-POMDP dialogue strategy outperforms three simple handcrafted dialogue strategies when the user's action error is induced by stress.

AB - We propose a new approach to developing a tractable affective dialogue model for general probabilistic frame-based dialogue systems. The dialogue model, based on the Partially Observable Markov Decision Process (POMDP) and the Dynamic Decision Network (DDN) techniques, is composed of two main parts, the slot level dialogue manager and the global dialogue manager. Our implemented dialogue manager prototype can handle hundreds of slots; each slot might have many values. A first evaluation of the slot level dialogue manager (1-slot case) showed that with a 95% confidence level the DDN-POMDP dialogue strategy outperforms three simple handcrafted dialogue strategies when the user's action error is induced by stress.

KW - HMI-MI: MULTIMODAL INTERACTIONS

KW - HMI-SLT: Speech and Language Technology

KW - METIS-241912

KW - EWI-8523

KW - IR-66715

M3 - Conference contribution

SN - not assigned

SP - 34

EP - 37

BT - Proceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems

PB - IJCAI

ER -

Bui Huu Trung BHT, Poel M, Nijholt A, Zwiers J. A tractable DDN-POMDP Approach to Affective Dialogue Modeling for General Probabilistic Frame-based Dialogue Systems. In Traum D, Alexandersson J, Jonsson A, Zukerman I, editors, Proceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems. Hyderabad, India: IJCAI. 2007. p. 34-37.