ELMo and BERT in semantic change detection for Russian
This work addresses semantic change detection for Russian, an incremental application of existing methods to a new language domain.
The study evaluated ELMo and BERT for detecting semantic change in Russian nouns and adjectives across pre-Soviet, Soviet, and post-Soviet periods, finding that these models effectively rank words by degree of change.
We study the effectiveness of contextualized embeddings for the task of diachronic semantic change detection for Russian language data. Evaluation test sets consist of Russian nouns and adjectives annotated based on their occurrences in texts created in pre-Soviet, Soviet and post-Soviet time periods. ELMo and BERT architectures are compared on the task of ranking Russian words according to the degree of their semantic change over time. We use several methods for aggregation of contextualized embeddings from these architectures and evaluate their performance. Finally, we compare unsupervised and supervised techniques in this task.