Bounds of the pos/neg scales for polarity

[chsuong]

Thanks for a great package, developers.

I'm trying to find out the boundaries for the polarity-based scores with the default option of sent_logit, ``a logit scale computed as $\mathrm{log}(pos + 0.5) - \mathrm{log}(neg + 0.5))$."

1. What are the boundaries of pos and neg in the scale, respectively? [0,5] and [-5,0]?

2. What would be the largest and smallest possible values of the polarity-based score for the example setting with the Lexicoder 2015 dictionary mentioned in the vignette?:

polarity(data_dictionary_LSD2015)
polarity(data_dictionary_LSD2015) <- 
  list(pos = c("positive", "neg_negative"), neg = c("negative", "neg_positive"))
print(data_dictionary_LSD2015, 0, 0)