A comparison of addressee detection methods for multiparty conversations

Rieks op den Akker, David Traum

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

    41 Downloads (Pure)

    Abstract

    Several algorithms have recently been proposed for recognizing addressees in a group conversational setting. These algorithms can rely on a variety of factors including previous conversational roles, gaze and type of dialogue act. Both statistical supervised machine learning algorithms as well as rule based methods have been developed. In this paper, we compare several algorithms developed for several different genres of muliparty dialogue, and propose a new synthesis algorithm that matches the performance of machine learning algorithms while maintaning the transparancy of semantically meaningfull rule-based algorithms.
    Original languageEnglish
    Title of host publicationProceedings of DiaHolmia 2009
    Subtitle of host publication13th Workshop on the Semantics and Pragmatics of Dialogue
    EditorsJens Edlund, Joakim Gustafson, Anna Hjalmarsson, Gabriel Skantze
    Place of PublicationStockholm
    PublisherKTH Royal Institute of Technology
    Pages99-106
    Number of pages8
    Publication statusPublished - 24 Jun 2009
    Event13th Workshop on the Semantics and Pragmatics of Dialogue, SemDial 2009: DiaHolmia - Sockholm, Sweden
    Duration: 24 Jun 200926 Jun 2009
    Conference number: 13

    Workshop

    Workshop13th Workshop on the Semantics and Pragmatics of Dialogue, SemDial 2009
    Abbreviated titleSemDial
    Country/TerritorySweden
    CitySockholm
    Period24/06/0926/06/09

    Keywords

    • METIS-275560
    • Machine classification
    • IR-75327
    • EWI-17164
    • EC Grant Agreement nr.: FP6/0033812
    • HMI-MI: MULTIMODAL INTERACTIONS
    • multimodal recognition machine classification
    • Multimodal recognition

    Fingerprint

    Dive into the research topics of 'A comparison of addressee detection methods for multiparty conversations'. Together they form a unique fingerprint.

    Cite this