Topical Text Classification of Russian News: a Comparison of BERT and Standard Models

The paper is devoted to the single-label topical classification of Russian news. The author compares the BERT features and standard character, word and structure-level features as text models. Experiments with OpenCorpora and eight news topics show that the BERT model is superior to standard ones, and achieves good classification quality for a small dataset of long news. Error analysis reveals the best classified topics: "economics", "culture", and "media". Comparison with the state-of-the-art research allows to consider BERT as a baseline for future investigations of analysis of texts in Russian.