Towards Automatic Identification of Discourse Markers in Dialogs: the Case of 'Like'

Zufferey, Sandrine; Popescu-Belis, Andrei (2004). Towards Automatic Identification of Discourse Markers in Dialogs: the Case of 'Like'. In: Proceedings of SIGDIAL'04 (5th SIGdial Workshop on Discourse and Dialogue). Cambridge, Massachussets. 30.04.-01.05.2004.

[img] Text
W04-2313.pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (116kB) | Request a copy

This article discusses the detection of discourse markers (DM) in dialog transcriptions, by human annotators and by automated means. After a theoretical discussion of the definition of DMs and their relevance to natural language processing, we focus on the role of like as a DM. Results from experiments with human annotators show that detection of DMs is a difficult but reliable task, which requires prosodic information from soundtracks. Then, several types of features are defined for automatic disambiguation of like: collocations, part-of-speech tags and duration-based features. Decision-tree learning shows that for like, nearly 70% precision can be reached, with near 100% recall, mainly using collocation filters. Similar results hold for well, with about 91% precision at 100% recall.

Item Type:

Conference or Workshop Item (Paper)

Division/Institute:

06 Faculty of Humanities > Department of Linguistics and Literary Studies > Institute of French Language and Literature

UniBE Contributor:

Zufferey, Sandrine

Subjects:

800 Literature, rhetoric & criticism > 840 French & related literatures
400 Language > 440 French & related languages

Language:

English

Submitter:

Sandrine Zufferey

Date Deposited:

25 Apr 2016 12:01

Last Modified:

25 Apr 2016 12:01

BORIS DOI:

10.7892/boris.78686

URI:

https://boris.unibe.ch/id/eprint/78686

Actions (login required)

Edit item Edit item
Provide Feedback