README.md

metadata

license: cc-by-nc-sa-4.0
dataset_info:
  features:
    - name: id
      dtype: string
    - name: text
      dtype: string
    - name: image
      dtype: image
    - name: img_path
      dtype: string
    - name: prop_label
      dtype:
        class_label:
          names:
            '0': not_propaganda
            '1': propaganda
    - name: hate_label
      dtype:
        class_label:
          names:
            '0': not-hateful
            '1': hateful
    - name: hate_fine_grained_label
      dtype:
        class_label:
          names:
            '0': sarcasm
            '1': humor
            '2': inciting_violence
            '3': mocking
            '4': other
            '5': exclusion
            '6': dehumanizing
            '7': contempt
            '8': inferiority
            '9': slurs
  splits:
    - name: train
      num_bytes: 156541594.307
      num_examples: 2143
    - name: dev
      num_bytes: 21725452
      num_examples: 312
    - name: test
      num_bytes: 45373687
      num_examples: 606
  download_size: 221704545
  dataset_size: 223640733.307

Prop2Hate-Meme

This repository presents the first Arabic Prop2Hate-Meme dataset which explore the intersection of propaganda and hate in memes using a multi-agent LLM-based framework. We extend an existing propagandistic meme dataset by annotating it with fine- and coarse-grained hate speech labels, and provide baseline experiments to support future research.

Table of contents:

• Dataset
• Licensing
• Citation

---

Dataset

We adopted the ArMeme dataset for both fine- and coarse-grained hatefulness categorization. We preserved the original train, development, and test splits. While ArMeme was initially annotated with four labels, for this study we retained only the memes labeled as propaganda and not_propaganda. These were subsequently re-annotated with hatefulness categories. The data distribution is provided below.

---

📊 Dataset Statistics

🏋️‍♂️ Train Split

prop_label

• propaganda: 603
• not_propaganda: 1540

hate_label

• not-hateful: 1930
• hateful: 213

hate_fine_grained_label

• sarcasm: 105
• humor: 1815
• inciting violence: 13
• mocking: 133
• other: 10
• exclusion: 6
• dehumanizing: 12
• contempt: 38
• inferiority: 4
• slurs: 7

---

🧪 Dev Split

prop_label

• not_propaganda: 224
• propaganda: 88

hate_label

• not-hateful: 281
• hateful: 31

hate_fine_grained_label

• humor: 260
• sarcasm: 19
• mocking: 19
• contempt: 7
• other: 1
• dehumanizing: 2
• inferiority: 1
• slurs: 1
• inciting violence: 2

---

🧾 Dev-Test Split (dev_test)

prop_label

• not_propaganda: 436
• propaganda: 170

hate_label

• not-hateful: 452
• hateful: 154

hate_fine_grained_label

• humor: 334
• sarcasm: 118
• inciting violence: 12
• slurs: 29
• other: 20
• mocking: 49
• contempt: 25
• inferiority: 14
• dehumanizing: 2
• exclusion: 3

---

Experimental Scripts

Please find the experimental scripts here: https://github.com/firojalam/propaganda-and-hateful-memes.git

Licensing

This dataset is licensed under CC BY-NC-SA 4.0. To view a copy of this license, visit https://creativecommons.org/licenses/by-nc-sa/4.0/

Citation

If you use our dataset in a scientific publication, we would appreciate using the following citations:

@inproceedings{alam2024propaganda,
  title={Propaganda to Hate: A Multimodal Analysis of Arabic Memes with Multi-agent LLMs},
  author={Alam, Firoj and Biswas, Md Rafiul and Shah, Uzair and Zaghouani, Wajdi and Mikros, Georgios},
  booktitle={International Conference on Web Information Systems Engineering},
  pages={380--390},
  year={2024},
  organization={Springer}
}

@inproceedings{alam2024armeme,
  title={{ArMeme}: Propagandistic Content in Arabic Memes},
  author={Alam, Firoj and Hasnat, Abul and Ahmed, Fatema and Hasan, Md Arid and Hasanain, Maram},
  booktitle={Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  year={2024},
  address={Miami, Florida},
  month={November 12--16},
  publisher={Association for Computational Linguistics},
}