I’m enjoying the fastai course, but currently lack the expertise to handle this ML problem at work. They’re using it to demonstrate the utility of old-school ML, and I was hoping to take a crack at it with the fastai stack. Please do let me know if there’s a more appropriate forum. I have tech support ticket data in the following format:
–5 free text fields. The first is roughly sentence-length, the second paragraph-length, the rest short.
–10 categorical (nominal) fields. Notably, each of these can take on one or more values: Both (Rochester) and (Miami | San Diego | Des Moines) would be valid.
–A random key associated with each example
–For the training set, a category label (A-E). This is to be determined for the test set.
I know that lessons 4 and 10 specifically address NLP, but I’ve only finished lesson 1 so far. I did read through the lesson 10 notes. I’d very much appreciate any guidance on how to attack this, or pointers towards relevant resources. Thank you!