INTEGRATION OF COMPUTATIONAL LINGUISTICS AND NLP IN TEACHING SYNONYMS

This article examines mechanisms for training synonym knowledge in NLP systems at the intersection of language pedagogy and computational linguistics. The aim is to propose a staged model to detect, cluster, and apply synonymy in educational tasks. The methodology integrates corpus-based distributional semantics, contextual embeddings, and supervised learning. Scientific novelty lies in linking synonymy evaluation criteria to didactic proficiency levels through an explicit training loop.