Mukaffi28/Uddessho-Bangla-Multimodal-Intent-Classification

📊 Uddessho Dataset — Multimodal Author Intent Classification

Uddessho (meaning "Intent" in English) is a multimodal dataset created for author intent classification in the low-resource Bangla language.
It contains 3,048 social media posts (text + images) labeled into six distinct intent types.

---

🏷️ Intent Categories & Label Mapping

Label ID	Class Name
0	Advocative
1	Controversial
2	Exhibitionist
3	Expressive
4	Informative
5	Promotive

---

📂 Dataset Overview

Split	Instances
Training	2,423
Testing	313
Validation	312
Total	3,048

---

📊 Distribution of Dataset Splits Across Intent Categories

Intent Category	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

---

📜 License

This dataset is released under the CC BY-SA 4.0 license. You are free to share and adapt the dataset for any purpose, including commercial use, as long as you provide appropriate credit and distribute any derivative works under the same license.

---

💻 Usage

from datasets import load_dataset

dataset = load_dataset("Mukaffi28/Uddessho-Bangla-Multimodal-Intent-Classification")
print(dataset["train"][0])

---

✏️ Citation

If you use the Uddessho Dataset in your research or project, please cite it as:

@InProceedings{10.1007/978-981-96-1758-6_32,
  author    = {Faria, Fatema Tuj Johora and Moin, Mukaffi Bin and Rahman, Md. Mahfuzur and Shanto, Md. Morshed Alam and Fahim, Asif Iftekher and Hoque, Md. Moinul},
  editor    = {Ullah, Abrar and Anwar, Sajid},
  title     = {Uddessho: An Extensive Benchmark Dataset for Multimodal Author Intent Classification in Low-Resource Bangla Language},
  booktitle = {Proceedings of International Conference on Information Technology and Applications},
  year      = {2025},
  publisher = {Springer Nature Singapore},
  address   = {Singapore},
  pages     = {383--393},
  isbn      = {978-981-96-1758-6}
}