Natural Language Processing
Chapter 4 · page 71
Definition
Natural language processing refers to a set of computational methods that allow a machine to process, interpret and produce human language, written or spoken. It is a field of research that has developed since the 1950s, at the crossroads of linguistics, computer science and, today, machine learning. Language is converted into usable data (words, sentences, structures), then analyzed by models in order to produce coherent results from the patterns detected.
Example
When you use a search engine, the system interprets your query, identifies the important words and sometimes infers your intent in order to offer relevant results. In the same way, machine translation tools and chatbots such as ChatGPT and Claude analyze sentence structure to produce a response suited to the context. These systems do not work on isolated words: they take in the whole sentence and its surroundings.
Why it matters
Natural language processing has become a central interface between humans and machines. It allows for more direct interaction, without going through technical commands. But language remains ambiguous, contextual and culturally situated, which limits performance: models can produce plausible answers without really grasping meaning, which explains certain errors and misunderstandings.
How to cite this text
Vial, S. (2026). Natural Language Processing. In A Living Lexicon of Artificial Intelligence (p. 71). Stéphane Vial, publisher, Montréal.
