Evons

Evons数据集是由哥伦比亚大学创建的，包含92,969篇来自真实和虚假新闻媒体的文章。该数据集特别之处在于每篇文章都附带了Facebook互动次数，作为文章传播性的指标。此外，还提供了文章描述和分享到Facebook时的缩略图，这些图像自动标注了对象标签和颜色属性。数据集主要用于分析和预测新闻的传播性，特别是在社交媒体上的影响。创建过程中，使用了多个自动化的图像分析工具来标注图像内容，确保数据集的质量和可用性。

The Evons Dataset was developed by Columbia University and contains 92,969 articles from both genuine and fake news outlets. A prominent feature of this dataset is that each article is paired with the count of Facebook interactions, serving as a metric for the article's spreadability. Additionally, article descriptions and thumbnails generated when sharing the articles to Facebook are provided, with these images automatically annotated with object tags and color attributes. This dataset is primarily used for analyzing and predicting the spreadability of news, particularly its impact on social media. During its creation, multiple automated image analysis tools were utilized to annotate the image content, ensuring the quality and usability of the dataset.