RBE_Sent Dataset Description:
The RBE_Sent (Roman Bengali-English Sentiment) dataset is a synthetic, gold-standard code-mixed dataset developed for sentiment analysis tasks involving Romanized Bengali and English. It captures real-world bilingual usage by blending Roman Bengali with English tokens within the same textual instances. The dataset is designed to support research in multilingual natural language processing, especially in the context of low-resource, code-mixed languages. Each… See the full description on the dataset page:
https://huggingface.co/datasets/DaliaBarua/RBE_Sent.
