Basic vocabulary of artificial intelligence
chatbot, deeply fake, great informationand so on. How to express these concepts in French?
To answer this question, DataFranca has just published a lexicon Artificial Intelligence 101 Words: A Quick Guide to Basic Data Science and Artificial Intelligence Vocabulary. Apart from translating key terms in the field, this book popularizes concepts to make them easier to understand.
For the scientific component, this project was directed by Patrick Drouin, professor of translation at the Faculty of Arts and Sciences of the University of Montreal, and Claude Coulomb, a physics and computer science graduate from UdeM.
It corresponds to the dictionary of artificial intelligence in French
Recognizing the rarity of terminology related to artificial intelligence, machine learning and deep learning in French, DataFranca, a non-profit organization, has collected more than 5,500 terms related to these fields on the wiki. The worldwide French-speaking community, composed of many scholars, terminologists and translators, has written thousands of encyclopedic entries submitted for peer review. Contributors to this project, supported by the Fonds de recherche du Québec and the Office québécois de la langue française, include Yoshua Bengio from Mila – the Quebec Institute for Artificial Intelligence – and Hugo Larochelle from Google.
Lexicon 101 words of artificial intelligence represents the heart of this work. Featuring 101 entries in French and English, it provides the most commonly used terms, such as deep learning, decision tree or layer of neurons.
Claude Coulombe and Patrick Drouin worked on this lexicon in pairs: the first devoted himself to the technical aspect, the second took on the popularization and linguistic aspect. “Claude discussed the concepts, then I adapted them into understandable explanations,” says Patrick Drouin.
Required movement speed
Claude Coulombe says: “It’s important, the speed of reaction, flexibility, because if we don’t go fast enough, the English term will dominate and then it will be very difficult to go back and undo the habits.”
For example, we think of many English computer terms used in French data set. Today, linguists struggle to implement the equivalent of a “dataset”.
Create new terms in French
Some words are translated almost literally chatbot, displayed by the “chatbot” or deep fraud, translates as “deep fraud”.
In some cases, metaphors have been preserved, for example, as mines for expression data mining, translates as “data acquisition”.
In other cases, researchers have chosen to express the terms more clearly: ad convert became “self-aware neural network” or “self-aware network”, which refers to “high-performance, sequentially deep neural network architecture”. It uses the mechanism of attention, more specifically, attention to itself, to replace both relapses and curls. Moreover, the use of focus in machine translation stems from the pioneering work of MILA,” we read in the dictionary.
Developing vocabulary
This dictionary is subject to change. great information until now it was translated as “big data”. “Now we know that the prefix mega- comes from the Greek word meaning “million” and today we are talking about data that is no longer stored in megabytes. [108]but go beyond the terabyte [109]petabyte [1015]there is no point in talking about it great information. That’s why we need to talk more great informationPatrick Drouin says.
