YouTube Chatter: Understanding Online Comments Discourse on Misinformative and Political YouTube Videos
Aarash Heydari, Janny Zhang, Shaan Appel, Xinyi Wu, Gireeja Ranade

TL;DR
This study analyzes YouTube comments on political misinformation, revealing highly polarized channels generate significantly more engagement and demonstrating how comment features can predict video bias.
Contribution
It provides the first comparative analysis of comments on misinformation versus trustworthy political videos and introduces classifiers for bias prediction based on comment features.
Findings
Polarized channels generate 7.5x more comments per view.
Misinformative channels generate 10.42x more replies per view.
Simple classifiers can predict video bias from comment statistics.
Abstract
We conduct a preliminary analysis of comments on political YouTube content containing misinformation in comparison to comments on trustworthy or apolitical videos, labelling the bias and factual ratings of our channels according to Media Bias Fact Check where applicable. One of our most interesting discoveries is that especially-polarized or misinformative political channels (Left-Bias, Right-Bias, PragerU, Conspiracy-Pseudoscience, and Questionable Source) generate 7.5x more comments per view and 10.42x more replies per view than apolitical or Pro-Science channels; in particular, Conspiracy-Pseudoscience and Questionable Sources generate 8.3x more comments per view and 11.0x more replies per view than apolitical and Pro-Science channels. We also compared average thread lengths, average comment lengths, and profanity rates across channels, and present simple machine learning classifiers…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsMisinformation and Its Impacts · Hate Speech and Cyberbullying Detection · Media Influence and Politics
