Speakers
Description
This paper presents NeoLex, a multilingual electronic lexical resource dedicated to contemporary Chinese–French and Vietnamese–French neology. Based on diachronic media corpora (2015–2025) and pre-2015 reference lexicons, the project combines corpus linguistics, Natural Language Processing, and computational lexicography to detect emerging lexical units. Candidate neologisms were extracted using Out-of-Vocabulary detection, Vietnamese n-gram extraction, and a character-based method for Chinese, then manually validated according to lexicographic criteria. The validated entries were modelled using OntoLex-Lemon, SKOS, and vartrans to represent lexical entries, lexical senses, translation equivalents between Chinese–French and Vietnamese–French, semantic domains, and diachronic metadata. A Streamlit-based prototype was developed to explore and visualize the resulting resources. NeoLex demonstrates how corpus-based neology detection and ontology-driven lexicography can contribute to interoperable multilingual electronic dictionaries.