Ce qui est dedans et ce qui est dehors

• Natural Language Processing
• Natural Language Understanding

Now say that A has invented a new device, say a coconut catapult. She excitedly sends detailed instructions on building a coconut catapult to B, and asks about B’s experiences and suggestions for improvements. Even if O had a way of construct- ing the catapult underwater, he does not know what words such as rope and coconut refer to, and thus can’t physically reproduce the experiment.Bender, Emily M., and Alexander Koller. “Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data.” In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, edited by Dan Jurafsky, Joyce Chai, Natalie Schluter, and Joel Tetreault, 5185–98. Online: Association for Computational Linguistics, 2020. https://doi.org/10.18653/v1/2020.acl-main.463.

We take meaning to be the relation M ⊆ E × I which contains pairs (e, i) of natural language expressions e and the communicative intents i they can be used to evoke. Given this definition of meaning, we can now use understand to refer to the process of retrieving i given e. Communicative intents are about something that is outside of language. When we say Open the window! or When was Malala Yousafzai born?, the communicative intent is grounded in the real world the speaker and listener inhabit together. Bender, Emily M., and Alexander Koller. “Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data.” In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, edited by Dan Jurafsky, Joyce Chai, Natalie Schluter, and Joel Tetreault, 5185–98. Online: Association for Computational Linguistics, 2020. https://doi.org/10.18653/v1/2020.acl-main.463.

Notre définition de la langue suppose que nous en écartons tout ce qui est étranger à son organisme, à son système, en un mot tout ce qu’on désigne par le terme de « linguistique externe ». Cette linguistique-là s’occupe pourtant de choses importantes, et c’est surtout à elles que l’on pense quand on aborde l’étude du langage. De Saussure, Cours de linguistique générale, 1916, Introduction Ch. V.

Il n’y a pas de hors texte

• Les frontières du système déterminent la définition de ce qui est modélisé.
• La matérialité de l’abstraction

• Besoin d’un système inductif?

• Besoin d’un modèle de langue?

• Besoin d’un chatbot?

• Des systèmes clos simples:

  • correspondance entre deux documents
  • règles explicites -> frontières claires