Terminology Research into the field of Deep Learning for Cosmography


Deep learning has sparked global interest in recent years within Artificial Intelligence (AI) and has been widely adopted in the fields of image recognition, speech recognition, natural language processing, and cosmology.

As a part of her ILTS Professional Master’s Degree (French acronym for Language Industry and Specialized Translation) at the Paris Diderot University, Julia Pagès wrote her Master’s thesis on a scientific article—chosen for terminology and translation research—entitled Measuring photometric redshifts using galaxy images and Deep Neural Networks. In July 2016, Ben Hoyle published this paper in the scientific journal Astronomy and Computing. Written in French, Julia’s work titled Utilisation de méthodes de deep learning pour la mesure du redshift photométrique consists of three parts: documentary research, terminology research, and the French translation of the article.

She analysed an article that stands at the intersection of three specialized fields—cosmography, photometry and deep learning. Cosmography is the science that maps the general features of the Universe by producing dynamic cartographies of the space around us. The questions of distance measurement and astronomical surveying are at the core of cosmology. This fundamental knowledge shapes our representation of the Universe and, therefore, the paradigm in which we operate. To measure the distance between the Earth and other galaxies, photometry can be combined with deep learning by using machine learning algorithms to train neural networks.

The latest multidisciplinary research generates a great deal of articles peppered with neologisms from various emerging domains like AI. With the aim of assisting translators confronted with similar scientific articles, her terminology research focuses on the study of terms extracted from the article and their collocations. She also provides a section about the translation of the article, which includes several translation versions for all segments.

You can read Julia Pagès Master’s Thesis here.

If you are interested in other theses and papers on terminology and linguistics, check out our Theses and Papers section.


Hoyle, Ben, Measuring photometric redshifts using galaxy images and Deep Neural Networks, Astronomy and Computing, Volume 16, July 2016, Pages 34-40, ISSN 2213-1337,


Written by Julia Pagès – Trainee at the Terminology Coordination Unit of the European Parliament (Luxembourg) and student of the LSCT Research Masters at the Paris Diderot University. She holds an ILTS Masters Degree (French acronym for Language Industry and Specialised Translation).