Skip to content

Latest commit

 

History

History
26 lines (20 loc) · 1.33 KB

README.md

File metadata and controls

26 lines (20 loc) · 1.33 KB
Raw

Description of files

snopes.csv and politifact.csv

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.

Ledwich_channelList.csv and Riberio_channelList.csv

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

videos.csv

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

Citation

@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"
}