• Aucun résultat trouvé

Our digital resource of 88,000 anonymised French text messages, the

N/A
N/A
Protected

Academic year: 2021

Partager "Our digital resource of 88,000 anonymised French text messages, the"

Copied!
1
0
0

Texte intégral

(1)

PLIN 2016: “Language and the new (instant) media” 12 May 2016, Louvain-la-Neuve, Belgium.

Domain Keywords: Mediated discourse analysis, Normalisation, Natural Language Processing.

Medium Keywords: SMS.

Cédric Lopez, Mathieu Roche, Rachel Panckhurst

“Non-standard texts: from theoretical positions to Natural Language Processing normalisation”

[50 words]

Our digital resource of 88,000 anonymised French text messages, the 88milSMS corpus, and sociolinguistic questionnaire data, are available (http://88milsms.huma-num.fr). Our theoretical position and Natural Language Processing (NLP) investigation techniques, including mediated discourse analysis on SMS-writing, ‘unknown’ item classification, alignment and normalisation methods, are envisaged for future implementation in real-life applications.

Références

Documents relatifs

– Our Requirement-to-Design Change Impact Analysis (RD-CIA) solution, marked (5) in Figure 1.1, focuses on propagating changes from NL requirements to system designs, when

Figure 3: Learning of correlations between terms extracted from the narratives and Unwanted Accident field values The mechanism consists in detecting how many times a

This study and others also show powerfully the nexus of social practices by which individuals build their social identities, impute identities to others or renegotiate the

Susanna Pechuro describes a rally called on by the democratic opposition to protest the emergency session of the Russian Republic's Congress of People's Deputies, organized to

The earliest claim to the existence of multiple kinds of discourse structure was made by Grosz and Sidner [10], who posited a linguistic structure (signalled.. by discourse cues),

Following this idea, we can hypothesize that, in our experiment, the relationship between action and language disappeared when the agent was a robot because French

Manon Cassier, Julien Longhi, Damien Nouvel, Agata Jackiewicz, Jean-Yves Antoine, Anaïs Lefeuvre-Halftermeyer1. To cite

In spite of the increasingly large textual datasets humanities researchers are confronted with, and the need for automatic tools to extract information from them,