Données de réplication pour : MALANTIN

The MALANTIN projetc (MAchine Learning Analysis for Non-Technological INnovation) is part of the projetc of the European distributed research infrastructure on innovation and researche (RISIS). This work presents a hybrid methodology for creating a multilingual corpus designed to identify terms related to new or emerging concepts, such as non-technological innovation, which constitutes our case study. This corpus, intended to become a structured dataset, was developed to feed machine learning models and meet the specific needs of the humanities and social sciences (HSS). To achieve this, we combined advanced web scraping, semantic annotation, and thematic lexicon generation techniques.