christinacdl/clickbait_notclickbait_dataset
收藏Hugging Face2023-06-22 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/christinacdl/clickbait_notclickbait_dataset
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是一个用于文本分类任务的英文数据集,规模在10K到100K之间。数据集经过清理,去除了重复项,并仅保留了首次出现的文本。数据集被分为训练集、测试集和验证集,分割比例分别为0.2和0.2。训练集大小为43,802,测试集大小为8,760,验证集大小为2,191。数据集的标签定义为0表示非点击诱饵,1表示点击诱饵。
This is an English-language dataset for text classification tasks, with a total size ranging between 10,000 and 100,000 instances. The dataset has been cleaned, with duplicate entries removed and only the first-occurring texts retained. It is split into training, test, and validation subsets, with the split ratios for the test and validation sets set to 0.2 respectively. The specific sample sizes of the subsets are 43,802 for the training set, 8,760 for the test set, and 2,191 for the validation set. The label definitions are as follows: 0 denotes non-clickbait, while 1 represents clickbait.
提供机构:
christinacdl原始信息汇总
数据集概述
基本信息
- 许可证: Apache-2.0
- 任务类别: 文本分类
- 语言: 英语
- 数据集大小: 10K<n<100K
数据结构
- 标签定义:
- 0: 非点击诱饵
- 1: 点击诱饵
数据处理
- 去重: 数据集已清理重复项,保留首次出现的文本。
- 数据分割:
- 训练集与测试集分割比例: 0.2
- 测试集与验证集分割比例: 0.2
数据集大小
- 训练集大小: 43,802
- 测试集大小: 8,760
- 验证集大小: 2,191



