Startseite Linguistik & Semiotik 8 Discourse markers and dialogue act annotation for computational dialogue systems
Kapitel
Lizenziert
Nicht lizenziert Erfordert eine Authentifizierung

8 Discourse markers and dialogue act annotation for computational dialogue systems

  • Darinka Verdonik
Veröffentlichen auch Sie bei De Gruyter Brill
Discourse Markers in Interaction
Ein Kapitel aus dem Buch Discourse Markers in Interaction

Abstract

The aim of this study is to uncover the interaction between DM functions and dialogue segmentation. More precisely, we investigate to what extent DMs are used as autonomous dialogue acts or as part of a larger dialogue act. Dialogue acts are defined as speech acts following Austin’s definition of illocutionary force. We hypothesize that the dialogual behaviour of DMs is informative of the function they will fulfil in discourse. The aim of the present study is thus to answer the following research question: Which DMs in what types of contexts can perform as dialogue acts on their own? Authentic conversation data in Slovene is used to conduct the analysis. As a result of the analysis, six criteria are defined that help distinguish DMs that are dialogue acts on their own. The importance of contextual bondedness and non-verbal, especially prosodic, features of speech in the interpretation of dialogue acts and DMs is outlined. The findings show that the majority of DM usages are not dialogue acts on their own, and at least some of those that are represent borderline DM usages.

Abstract

The aim of this study is to uncover the interaction between DM functions and dialogue segmentation. More precisely, we investigate to what extent DMs are used as autonomous dialogue acts or as part of a larger dialogue act. Dialogue acts are defined as speech acts following Austin’s definition of illocutionary force. We hypothesize that the dialogual behaviour of DMs is informative of the function they will fulfil in discourse. The aim of the present study is thus to answer the following research question: Which DMs in what types of contexts can perform as dialogue acts on their own? Authentic conversation data in Slovene is used to conduct the analysis. As a result of the analysis, six criteria are defined that help distinguish DMs that are dialogue acts on their own. The importance of contextual bondedness and non-verbal, especially prosodic, features of speech in the interpretation of dialogue acts and DMs is outlined. The findings show that the majority of DM usages are not dialogue acts on their own, and at least some of those that are represent borderline DM usages.

Heruntergeladen am 2.10.2025 von https://www.degruyterbrill.com/document/doi/10.1515/9783110790351-008/html
Button zum nach oben scrollen