Transfer Learning for Improving Results on Russian Sentiment Datasets
This work addresses sentiment analysis for Russian language users, representing an incremental advance by adapting existing methods to a specific domain.
The study applied transfer learning with distant supervision to Russian sentiment datasets, achieving over 3% improvement on most datasets and reaching human-level performance on one dataset using a BERT-NLI model.
In this study, we test transfer learning approach on Russian sentiment benchmark datasets using additional train sample created with distant supervision technique. We compare several variants of combining additional data with benchmark train samples. The best results were achieved using three-step approach of sequential training on general, thematic and original train samples. For most datasets, the results were improved by more than 3% to the current state-of-the-art methods. The BERT-NLI model treating sentiment classification problem as a natural language inference task reached the human level of sentiment analysis on one of the datasets.