Skip to Main content Skip to Navigation
Master Thesis

Développement d'un module de génération de paraphrases pour la data augmentation

Abstract : This paper summarizes the work completed during a six-month internship in the Research & Development team of Linagora. The aim of this project is to develop a paraphrasing tool able to expand the training datasets of the smart vocal assistant LinTO. We started by analyzing the existing commands in the original corpus : the recurrent syntactic structures and the paraphrasing mecanisms that can be applied to them. From these observations, we created a rule-based grammar to generate semantically and synctactically correct sentences.
Document type :
Master Thesis
Complete list of metadatas

Cited literature [24 references]  Display  Hide  Download

https://dumas.ccsd.cnrs.fr/dumas-02294883
Contributor : Uga - Bulles <>
Submitted on : Monday, September 23, 2019 - 5:51:42 PM
Last modification on : Wednesday, July 15, 2020 - 1:06:02 PM
Long-term archiving on: : Sunday, February 9, 2020 - 8:48:26 AM

File

RATSIANDAVANA Sonia_M2_IDL.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : dumas-02294883, version 1

Citation

Sonia Ratsiandavana. Développement d'un module de génération de paraphrases pour la data augmentation. Sciences de l'Homme et Société. 2019. ⟨dumas-02294883⟩

Share

Metrics

Record views

40

Files downloads

21