This module starts from a list of seeds and crawls networks of channels connected via the “related channels” panel which YouTube generates algorithmically. (the channel network module uses featured channels and subscriptions as connections, which are set byt the channel owner).

Related channel data are not retrieved via the API, but scraped from YouTube’s web interface. To reduce the number of requests, this module caches channel data for three days. Remember that channels can opt out of this feature and will thus not appear in the network.

Crawl depth specifies how far from the seeds the script should go. Crawl depth 0 will get only the relations between seeds. Using many seeds and the maximum crawl depth (3) can take a very long time or the script might run out of memory or get blocked. Start small.

NB: since graph analysis software can have difficulties with very large numbers, channels’ viewcount is given in 100s.