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

Trung H. Bui; Mannes Poel; Anton Nijholt; Job Zwiers

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

Trung H. Bui, Mannes Poel, Anton Nijholt, Job Zwiers

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

75 Downloads (Pure)

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 language	English
Title of host publication	Proceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems
Editors	David Traum, Jan Alexandersson, Arne Jönsson, Ingrid Zukerman
Place of Publication	Hyderabad, India
Publisher	IJCAI
Pages	34-37
Number of pages	4
Publication status	Published - 8 Jan 2007
Event	5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems 2007 - Hyderabad, India Duration: 8 Jan 2007 → 8 Jan 2007

Workshop

Workshop	5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems 2007
Country/Territory	India
City	Hyderabad
Period	8/01/07 → 8/01/07

Keywords

HMI-MI: MULTIMODAL INTERACTIONS
HMI-SLT: Speech and Language Technology

Access to Document

Bui2006tractableFinal published version, 283 KB

1 Report

A tractable DDN-POMDP Approach to Affective Dialogue Modeling for General Probabilistic Frame-based Dialogue Systems
Bui, T. H., Poel, M., Nijholt, A. & Zwiers, J., 28 Sept 2006, Enschede: Centre for Telematics and Information Technology (CTIT). 8 p. (CTIT Technical Report Series; no. 06-65)
Research output: Book/Report › Report › Professional

Open Access
File

Cite this

Bui, Trung H. ; Poel, Mannes ; Nijholt, Anton et al. / 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. editor / David Traum ; Jan Alexandersson ; Arne Jönsson ; Ingrid Zukerman. Hyderabad, India : IJCAI, 2007. pp. 34-37

@inproceedings{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",

author = "Bui, {Trung H.} and Mannes Poel and Anton Nijholt and Job Zwiers",

year = "2007",

month = jan,

day = "8",

language = "English",

pages = "34--37",

editor = "David Traum and Jan Alexandersson and Arne J{\"o}nsson and Ingrid Zukerman",

booktitle = "Proceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems",

publisher = "IJCAI",

note = "5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems 2007 ; Conference date: 08-01-2007 Through 08-01-2007",

}

Bui, TH, 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 Jönsson & I Zukerman (eds), Proceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems. IJCAI, Hyderabad, India, pp. 34-37, 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems 2007, Hyderabad, India, 8/01/07.

A tractable DDN-POMDP Approach to Affective Dialogue Modeling for General Probabilistic Frame-based Dialogue Systems. / Bui, Trung H.; Poel, Mannes ; Nijholt, Anton et al.
Proceedings of the 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems. ed. / David Traum; Jan Alexandersson; Arne Jönsson; Ingrid Zukerman. Hyderabad, India: IJCAI, 2007. p. 34-37.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

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

AU - Bui, Trung H.

AU - Poel, Mannes

AU - Nijholt, Anton

AU - Zwiers, Job

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

M3 - Conference contribution

SP - 34

EP - 37

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

A2 - Traum, David

A2 - Alexandersson, Jan

A2 - Jönsson, Arne

A2 - Zukerman, Ingrid

PB - IJCAI

CY - Hyderabad, India

T2 - 5th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems 2007

Y2 - 8 January 2007 through 8 January 2007

ER -

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

Abstract

Workshop

Keywords

Access to Document

Fingerprint

Research output

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

Cite this