OffMix-3L: A Novel Code-Mixed Dataset in Bangla-English-Hindi for Offensive Language Identification
This work addresses a gap in NLP resources for multi-language code-mixing, specifically for offensive language identification, but is incremental as it builds on existing datasets and tasks.
The authors tackled the lack of datasets for code-mixed data involving three languages by introducing OffMix-3L, a novel dataset for offensive language identification in Bangla-English-Hindi, and found that BanglishBERT outperformed other models like GPT-3.5 on this dataset.
Code-mixing is a well-studied linguistic phenomenon when two or more languages are mixed in text or speech. Several works have been conducted on building datasets and performing downstream NLP tasks on code-mixed data. Although it is not uncommon to observe code-mixing of three or more languages, most available datasets in this domain contain code-mixed data from only two languages. In this paper, we introduce OffMix-3L, a novel offensive language identification dataset containing code-mixed data from three different languages. We experiment with several models on this dataset and observe that BanglishBERT outperforms other transformer-based models and GPT-3.5.