This is outside of my area of expertise, but I wonder whether it'd be possible to train machine learning models to identify divergent threads of conversation.

This is outside of my area of expertise, but I wonder whether it'd be possible to train machine learning models to identify divergent threads of conversation. Tangential conversation can be useful and constructive sometimes; but other times it'd be nice to apply a contextual overlay to a conversation which emboldens the on-topic discussion items and de-prioritizes side discussions. It's possible to argue that that's what the HN community itself should do naturally via upvote/downvote behaviour -- but in practice there are various (often valid) reasons why an unrelated thread of commentary can become more popular than on-topic discussion.