These two files contain "fake news" headlines as reported by the fact checking websites snopes.com and politifact.com. These headlines were used in our "fake news" analysis to compute the proportion of videos that discuss topics deemed to be fake news.
These two files contain the channelID's of the channels collected from previous research on the YouTube recommendation system, namely these two studies:
Riberio et al.: https://arxiv.org/abs/1908.08313
Ledwich et al.: https://firstmonday.org/ojs/index.php/fm/article/view/10419/9404
This file contains all videos in our dataset. Since it was too large to upload to GitHub, it can be found here instead: https://drive.google.com/file/d/1zDBPGdu_jl0eV1vdy3GwB7JpEFvrKnoE/view?usp=sharing
@article{10.1093/pnasnexus/pgad264,
author = "Ibrahim, Hazem and AlDahoul, Nouar and Lee, Sangjin and Rahwan, Talal and Zaki, Yasir",
title = "{YouTube's recommendation algorithm is left-leaning in the United States}",
journal = "PNAS Nexus",
pages = "pgad264",
doi = "10.1093/pnasnexus/pgad264",
url = "https://doi.org/10.1093/pnasnexus/pgad264"
year = "2023"
}