arbml/MLMA_hate_speech_ar

该数据集包含社交媒体上的推文数据，每条推文都有唯一的HITId标识，推文内容存储在tweet字段中。sentiment字段表示推文的情感类别，共有36种不同的情感标签，包括abusive、disrespectful、hateful、normal等。directness字段表示推文的直接性，annotator_sentiment字段表示标注者的情感判断，target字段表示推文的目标对象，group字段表示推文所属的群体。数据集被分为训练集，包含3353个样本。

This dataset is primarily used for sentiment analysis, containing multiple features such as HITId, tweet, sentiment, etc. The sentiment feature includes multiple class labels to represent different types of sentiments, such as abusive, disrespectful, fearful, etc. The dataset is divided into a training set with 3353 samples, and the dataset size is 641818 bytes, with a download size of 278807 bytes.