Feature Request: Part of speech tagging

[benmarwick]

This would be great, the R packages are slow and haven't been updated for years...

[johnmyleswhite]

I'll try to get to this in early June.

[benmarwick]

Sounds like you've got a busy summer plan! Thanks for all you've done so far with this, I'm looking forward to the LDA function also. Unfortunately pretty much all I can offer is testing...

[johnmyleswhite]

Testing is definitely helpful. I'm hoping to get some LDA input from Jon Chang and David Mimno in the future.

[cgoldammer]

Hi John, I assume that this is still relevant? I've found a promising POS tagger here: https://honnibal.wordpress.com/2013/09/11/a-good-part-of-speechpos-tagger-in-about-200-lines-of-python/ that should be relatively straightforward to translate to Julia. I'm aiming to code this up over the next weeks. If it works well enough, I'd love to put it in a pull request.

[leoarc]

@aviks @oxinabox Is the above comment still relevant ? Should the link given be followed and implemented for a POS tagger ?

[oxinabox]

At a skim, yes, that looks like a good guide for writing a POS tagger.

[Ayushk4]

@aviks I think we can close this issue.
In #131, YashPatel implemented the avg perceptron POS Tagger.

Also, a neural network based POS tagger was added in #169 that gives >90% accuracy on CoNLL 2003, 2000, GMB. (link)