When building chatbots with NLU (Natural Language Understanding), we need a lot of training samples to train intent detection system. One way is to collect data from users by extracting conversations or using services like Amazon Mechanical Turk, but this is tedious work or costly processes (in addition to the fact that writing sentences by hand is not a really fun job). The other way is to develop a language model that is capable of paraphrasing sentences. A paraphrase is a restatement of a text using other words. In this way, it will be possible to create a fraction of the sentences manually and then generate tens of new sentences based on them by paraphrasing them.

General information
  • Date: 05.02.2021
  • Type: Master project
  • Responsible: Karl Daher


  • Samuel Torche
Jacky Casas
Omar Abou Khaled
See more
Karl Daher
Senior Researcher
See more
Retus Muller
Partners : HEIA-FR   Humantech Institute