SPIRS is a unique dataset of 15,000 sarcastic tweets, together with 15,000 non-sarcastic tweets for a total of 30K samples. SPIRS was collected using reactive supervision, a new data capturing method. Reactive supervision allows the collection of both intended sarcasm and perceived sarcasm texts.
SPIRS stands for Sarcasm, Perceived and Intended, by Reactive Supervision :)
To find out more about SPIRS and reactive supervision, check out the reactive supervision paper, or read the Medium article. Or watch this short, 7-minute YouTube video about reactive supervision.
Use this repository to download SPIRS. The repository includes the following data files:
SPIRS-sarcastic-ids.csvthe sarcastic tweet IDs (15,000 "positive" samples)SPIRS-non-sarcastic-ids.csvthe non-sarcastic tweet IDs (15,000 "negative" samples)
Additional fields for each sarcastic tweet include the sarcasm perspective (intended/perceived), author sequence, and contextual tweet IDs (cue, oblivious, and eliciting tweets). More information is available in the reactive supervision paper.
To comply with Twitter's privacy policy, the dataset files include only the tweet IDs. To fetch the tweet texts, follow these steps:
-
Install the latest version of Tweepy:
pip3 install tweepy -
Rename our
credentials-example.pytocredentials.py -
Add your Twitter API credentials by editing
credentials.py -
Run the script:
python3 fetch-tweets.py
The script will fetch the texts and create two new files, one for sarcastic and the other for non-sarcastic tweets:
SPIRS-sarcastic.csvSPIRS-non-sarcastic.csv
Kindly cite the paper using the following BibTex entry:
@inproceedings{shmueli-etal-2020-reactive,
title = "{R}eactive {S}upervision: {A} {N}ew {M}ethod for {C}ollecting {S}arcasm {D}ata",
author = "Shmueli, Boaz and
Ku, Lun-Wei and
Ray, Soumya",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.emnlp-main.201",
doi = "10.18653/v1/2020.emnlp-main.201",
pages = "2553--2559",
abstract = "Sarcasm detection is an important task in affective computing, requiring large amounts of labeled data. We introduce reactive supervision, a novel data collection method that utilizes the dynamics of online conversations to overcome the limitations of existing data collection techniques. We use the new method to create and release a first-of-its-kind large dataset of tweets with sarcasm perspective labels and new contextual features. The dataset is expected to advance sarcasm detection research. Our method can be adapted to other affective computing domains, thus opening up new research opportunities.",
}