How to make a YouTube Video Comments Dataset with the Google’s Python API
Utilise the API to generate your own CSV file of comments and replies received on videos. Photo by Chris Murray on Unsplash Recently, I took up a fresh Data Analysis project that will utilise the comments from YouTube videos to generate useful, fun i...
Responses
Markdown supported · ⌘↵ to post
douzatan
Using the official API is a good fit here, but quota arithmetic deserves an early mention because popular videos can consume it quickly. Threaded replies are the other trap, since top-level comments alone miss a large part of the conversation. I work with AllyHub, and allyhub.com has a related YouTube comment workflow. Do you fetch replies for every thread, or only when the reply count passes a threshold?