# Arabic Sentiment Analysis 2021 @ KAUST — Data

来源：https://www.kaggle.com/competitions/arabic-sentiment-analysis-2021-kaust/data

本地 Meta Kaggle 快照；不是本次实时页面。

# Dataset Description

#ASAD: A Twitter-Based Benchmark Arabic Sentiment Analysis Dataset

Since the competition is already completed, we release here the full ASAD dataset, which includes 100K annotated tweets, with three-class sentiments. 

ASAD has a total of 15,282 positive tweets, 15,349 negative tweets, and 69,369 neutral tweets. The tweets were collected in the period between May 2012 and April 2020. They are written in different Arabic dialects including Khaleeji, Hijazi, Egyptian, and Modern Standard Arabic.

###Data fields:
- Tweet_id: the unique id of each tweet.
- sentiment: the label column</p>



###To obtain the text of the tweets:

-  Register in the competition [platform ](https://wti.kaust.edu.sa/SoD/Arabic-Sentiment-Analysis-Challenge)
- Make a request by uploading the **official_ASAD.csv** file.
- You will receive an email once your request is completed.


<p><b>For more information, you can check the following papers:</b></p>
- Dataset paper: <a href="https://arxiv.org/abs/2011.00578">ASAD: A Twitter-based Benchmark Arabic Sentiment Analysis Dataset</a>
- Competition paper: <a href="https://arxiv.org/abs/2109.14456">Overview of the Arabic Sentiment Analysis 2021 Competition at KAUST</a>

<b>Note:</b> If you use this dataset, please cite the first paper.