Uddessho: An Extensive Benchmark Dataset for Multimodal Author Intent Classification in Low-Resource Bangla Language
Description
The "Uddessho" dataset, meaning "Intent" in English, is designed for multimodal author intent classification. It contains 3048 post instances categorized into six intent types: Informative, Advocative, Promotive, Exhibitionist, Expressive, and Controversial. The dataset is divided into a training set with 2423 posts, a testing set with 313 posts, and a validation set with 312 posts, totaling 3048 posts. Distribution of Dataset Splits Across Different Intent Categories ========================================= Intent Taxonomy Training Testing Validation ========================================= Informative 514 67 67 Advocative 386 49 49 Promotive 315 43 42 Exhibitionist 371 47 48 Expressive 518 66 66 Controversial 319 41 40 ========================================= Total 2423 313 312 =========================================
Files
Institutions
- Ahsanullah University of Science and Technology