VLEngagement

VLEngagement数据集由伦敦大学学院人工智能中心创建，包含超过4000个经过同行评审的科学视频讲座，旨在评估群体基础的参与度。数据集涵盖了文本和视频特定的特征，以及与用户参与度相关的多种指标。创建过程涉及从公开可用的科学视频讲座中提取内容相关和视频特定的特征。该数据集特别适用于构建智能教育和科学推荐系统，解决教育推荐系统中的冷启动问题，并帮助理解教育材料中影响群体基础参与度的特征。

The VLEngagement dataset was created by the Centre for Artificial Intelligence at University College London (UCL). It contains over 4,000 peer-reviewed scientific video lectures, with the aim of evaluating group-based engagement. The dataset covers both text-related and video-specific features, as well as multiple metrics related to user engagement. The dataset's creation process involves extracting content-related and video-specific features from publicly available scientific video lectures. This dataset is particularly suitable for developing intelligent educational and scientific recommendation systems, addressing the cold-start problem in educational recommendation systems, and aiding in the understanding of features that influence group-based engagement in educational materials.