ArSarcasm Dataset

ArSarcasm is a new Arabic sarcasm detection dataset. The dataset was created using previously available Arabic sentiment analysis datasets (SemEval 2017 and ASTD) and adds sarcasm and dialect labels to them. The dataset contains 10,547 tweets, 1,682 (16%) of which are sarcastic. For more details, please check our paper From Arabic Sentiment Analysis to Sarcasm Detection: The ArSarcasm Dataset

Dataset details:

ArSarcasm is provided in a CSV format, we provide an 80/20 train/test split to keep things consistent for future comparisons. The training set contains 8,437 tweets, while the test set contains 2,110 tweets.

The dataset contains the following fields:

• tweet: the original tweet text surrounded by quotes (").
• sarcasm: boolean that indicates whether a tweet is sarcastic or not.
• sentiment: the sentiment from the new annotation (positive, negative, neutral).
• original_sentiment: the sentiment in the original annotations (positive, negative, neutral).
• source: the original source of tweet SemEval or ASTD.
• dialect: the dialect used in the tweet, we used the 5 main regions in the Arab world, follows the labels and their meanings:
  • msa: modern standard Arabic.
  • egypt: the dialect of Egypt and Sudan.
  • levant: the Levantine dialect including Palestine, Jordan, Syria and Lebanon.
  • gulf: the Gulf countries including Saudi Arabia, UAE, Qatar, Bahrain, Yemen, Oman, Iraq and Kuwait.
  • magreb: the North African Arab countries including Algeria, Libya, Tunisia and Morocco.

Citation

Please use the following citation if you use ArSarcasm:

@inproceedings{abu-farha-magdy-2020-arabic,
    title = "From {A}rabic Sentiment Analysis to Sarcasm Detection: The {A}r{S}arcasm Dataset",
    author = "Abu Farha, Ibrahim  and Magdy, Walid",
    booktitle = "Proceedings of the 4th Workshop on Open-Source Arabic Corpora and Processing Tools, with a Shared Task on Offensive Language Detection",
    month = may,
    year = "2020",
    address = "Marseille, France",
    publisher = "European Language Resource Association",
    url = "https://www.aclweb.org/anthology/2020.osact-1.5",
    pages = "32--39",
    language = "English",
    ISBN = "979-10-95546-51-1",
}

Other resources

If you are interested in other Arabic NLP resources check:

• Mazajak Sentiment Analyser
• Mazajak Word Embeddings