Hacker News Reader

It's identifying the collocations so that you can do that replacement that remains an imperfect science.

I've heard good things about "Scalable Topical Phrase Mining from Text Corpora" [1], but it's been a while, so I don't know how close to the state of the art it is.

[1] https://arxiv.org/abs/1406.6312

Reply on news.ycombinator.com