Suggestions for section on "Creating Features from Text Data" about fes

[topepo]

textfeatures is a huge help. Any idea what n_tobe computes? The docs don't really say. At some point, I'd like to develop a bunch of text-oriented recipe steps.

Thanks for the edits, you are a much better writer than I.

On stemming and stop words: those articles are very convincing. I did the keyword selection without any word editing and came up with virtually the same results (nurs was dropped and another was picked up). I'll remove the extra processing and describe those steps closer to the end.

I'm also going to split out the types of features; the keyword effect is large (as expected):

[juliasilge]

textfeatures is really great but the documentation leaves a lot to be desired currently. Mike does have the repo marked "experimental" still? FWIW n_tobe measures "to be" verbs.

I would love for him to get this package matured because I would like to feel better about recommending it to folks.

[topepo]

Thanks for all of your help.