Affective dialogue management using factored POMDPs

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

    Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

    6 Citations (Scopus)

    Abstract

    Partially Observable Markov Decision Processes (POMDPs) have been demonstrated empirically to be good models for robust spoken dialogue design. This chapter shows that such models are also very appropriate for designing affective dialogue systems. We describe how to model affective dialogue systems using POMDPs and propose a novel approach to develop an affective dialogue model using factored POMDPs. We apply this model for a single-slot route navigation dialogue problem as a proof of concept. The experimental results demonstrate that integrating user’s affect into a POMDP-based dialogue manager is not only a nice idea but also helpful for improving the dialogue manager performance given that the user’s affect influences their behavior. Further, our practical findings and experiments on the model tractability are expected to be helpful for designers and researchers who are interested in practical implementation of dialogue systems using the state-of-the-art POMDP techniques.
    Original languageUndefined
    Title of host publicationInteractive Collaborative Information Systems
    EditorsR. Babuska, F.C.A. Groen
    Place of PublicationBerlin
    PublisherSpringer
    Pages209-238
    Number of pages30
    ISBN (Print)978-3-642-11687-2
    DOIs
    Publication statusPublished - 24 Mar 2010

    Publication series

    NameStudies in Computational Intelligence
    PublisherSpringer Verlag
    Number281
    Volume281
    ISSN (Print)1860-949X

    Keywords

    • EWI-15201
    • HMI-MI: MULTIMODAL INTERACTIONS
    • POMDPs
    • METIS-270686
    • Affective dialogue systems
    • Navigation
    • IR-70516

    Cite this