Practical Dialogue Manager Development using POMDPs

B.H.T. Bui Huu Trung, B.W. van Schooten, D.H.W. Hofs

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    5 Citations (Scopus)
    18 Downloads (Pure)

    Abstract

    Partially Observable Markov Decision Processes (POMDPs) are attractive for dialogue management because they are made to deal with noise and partial information. This paper addresses the problem of using them in a practical development cycle. We apply factored POMDP models to three applications. We examine our experiences with respect to design choices and issues, and compare performance with hand-crafted policies.
    Original languageUndefined
    Title of host publicationProceedings of the 8th SIGdial Workshop on Discourse and Dialogue
    EditorsS. Keizer, H. Bunt, T. Paek
    Place of PublicationPA, USA
    PublisherAssociation for Computational Linguistics (ACL)
    Pages215-218
    Number of pages4
    ISBN (Print)978-9-07402-932-2
    Publication statusPublished - 1 Sep 2007

    Publication series

    Name
    PublisherThe Association for Computational Linguistics
    Number2

    Keywords

    • EWI-9363
    • HMI-SLT: Speech and Language Technology
    • METIS-242046
    • IR-66968
    • HMI-MI: MULTIMODAL INTERACTIONS

    Cite this

    Bui Huu Trung, B. H. T., van Schooten, B. W., & Hofs, D. H. W. (2007). Practical Dialogue Manager Development using POMDPs. In S. Keizer, H. Bunt, & T. Paek (Eds.), Proceedings of the 8th SIGdial Workshop on Discourse and Dialogue (pp. 215-218). PA, USA: Association for Computational Linguistics (ACL).