Depression dataset for Arabic tweets Binary-classification

The dataset contains 10,000 Arabic tweets extracted from twitter as a depression dataset. The tweets classified as 5000 tweets labeled as depressed, and 5000tweets labeled as nondepressed. The labeling process is made manually by authors and medicals expert support and validation.

本数据集包含10000条从推特（Twitter）采集的阿拉伯语推文，为抑郁症相关研究数据集。所有推文被划分为两类：5000条标注为抑郁样本，5000条标注为非抑郁样本。标注工作由本文作者手动完成，并获得医学专家的支持与验证。