Leveraging Contextual Embeddings for Detecting Diachronic Semantic Shift
This addresses the problem of tracking how word meanings change over time for computational linguistics and NLP researchers, with incremental improvements in efficiency and applicability.
The authors tackled diachronic semantic shift detection by proposing a method using contextual embeddings from BERT to generate time-specific word representations, achieving performance comparable to state-of-the-art on the LiverpoolFC corpus without domain adaptation and showing effectiveness for short-term shifts on a Brexit news corpus and in multilingual settings.
We propose a new method that leverages contextual embeddings for the task of diachronic semantic shift detection by generating time specific word representations from BERT embeddings. The results of our experiments in the domain specific LiverpoolFC corpus suggest that the proposed method has performance comparable to the current state-of-the-art without requiring any time consuming domain adaptation on large corpora. The results on the newly created Brexit news corpus suggest that the method can be successfully used for the detection of a short-term yearly semantic shift. And lastly, the model also shows promising results in a multilingual settings, where the task was to detect differences and similarities between diachronic semantic shifts in different languages.