English is the language of reference among scientists. Dissemination in a dominant language, such as English, necessarily influences the production of knowledge and this can have a number of consequences especially in the evaluative metrics for scientific research. The fairness of scientific research evaluation depends to a large extent on the comprehensiveness and accuracy of citation data. English has a greater influence in the scientific field, it is inevitable that language bias will result in an unfair evaluation of the impact of scientific research indicators on other languages’ research. This makes English more and more dominant in knowledge dissemination, while gradually weakening the influence of other languages in terms of scientific development. However, while English has had a form of hegemony in recent decades, this has not always been the case. Russian, French and German, to name but a few, are still used to produce scientific concepts. Indeed, we think in our language and concepts vary with linguistic diversity and different cultures. Building science means building concepts and this is done through dialogue, controversy and scientific debate. The question that arises is: do we mobilise concepts identically from one language to another? Conversely, can the same notions and concepts be produced and defined identically if they are formulated in different languages? Within the context of this study, we are particularly interested in the study of citation acts and more particularly the study of citation contexts, which represents an interdisciplinary theme. Indeed, it requires knowledge in bibliometry but also in computer science and automatic language processing. The study of citation contexts implies the study of references present in the text and the study of their contexts. Numerous studies have been carried out in this area, attempting to establish a semantic dimension to the nature of the relations involved.