This is also on page 356.

from nltk.corpus import sentiwordnet as swn

good = swn.senti_synsets('good', 'n')[0]
Traceback (most recent call last):
File "", line 1, in
TypeError: 'filter' object is not subscriptable

In #1925, we noticed that normalization of hyphens in the corpus constructor is causing inconsistency.

https://github.com/quanteda/quanteda/blob/5a8b661944d4917006a23131bd850c3bee4be32c/R/corpus.R#L153-L156

If corpus is a an object to keep original texts, we probably should not do it in the corpus constructor but in token constructor.

OS: MacOS 10.13.3
I installed the MeTA as instructed on setup guide. When I do the unit test

describe [ranker regression]
libc++abi.dylib: terminating with uncaught exception of type meta::corpus::corpus_exception: corpus configuration file (../data//cranfield/line.toml) not present

I am playing around with MALLET and TMT, I see a lot of optional settings with default values, but I have not found explanations for how these settings affect results, and what best practices are for choosing them. I would very much appreciate a guide explaining what each setting does, and advice on how to choose a proper value or evaluate the difference between values.

As an example: why is th

In particular the user documentation needs to show users how to fetch queries from TAP, obtain the descriptions, and the authority information