Ask HN: Is machine learning useful outside the few very common applications?

In theory, we can use machine learning to do a lot of cool things. However, I've seen a dearth of material on how real world problems are formulated in terms of machine learning problems.

Computer Vision is rapidly advancing as far as deep learning is concerned, and we can generally see a lot of applications there (medical, autonomous, etc.)

NLP has a different story. For example, I was researching how a support desk uses machine learning. They use that to classify tickets automatically to the right department. Useful? Somewhat. Revolutionary? I don't think so. [this was from MonkeyLearn's website, btw]

What are the kind of people I can connect with who are doing a lot of applicative work when it comes to machine learning and AI? I mean, focusing on real-world problems and then formulating them as an ML problem? Or are there resources that I could read?

Thanks!

from Hacker News: Newest https://ift.tt/2trRYFC