fioriclass/suicide_prediction_dataset_phr

自杀倾向预测数据集，来源于Reddit，包含文本数据和表示自杀或非自杀的二元标签。数据集经过预处理，包括文本转换为小写、去除数字、特殊字符、URL、表情、重音符号、缩写、额外空格以及连续超过3次的字符。文本经过分词、词形还原和停用词去除。数据集分为训练集和测试集，比例为80:10:10。

Suicidal Tendency Prediction Dataset sourced from Reddit, containing text data and binary labels indicating suicide or non-suicide. The dataset has been preprocessed, including converting text to lowercase, removing numbers, special characters, URLs, emojis, accented characters, contractions, extra spaces, and consecutive characters repeated more than three times. The text has been tokenized, lemmatized, and stopwords removed. The dataset is split into training and test sets with a ratio of 80:10:10.