Closed cmaimone closed 2 years ago
@cmaimone I thought that S_diff
and T_diff
are not very useful for interpretation (unlike p
from other tests; although a more positive diff
of a word indicates a higher association with A
relative to B
and vice versa). But yeah, for the sake of completeness, the two vectors should also be named. It is now fixed (and tested).
require(sweater)
#> Loading required package: sweater
S4 <- c("math", "algebra", "geometry", "calculus", "equations", "computation", "numbers", "addition")
T4 <- c("poetry", "art", "dance", "literature", "novel", "symphony", "drama", "sculpture")
A4 <- c("male", "man", "boy", "brother", "he", "him", "his", "son")
B4 <- c("female", "woman", "girl", "sister", "she", "her", "hers", "daughter")
sw <- query(glove_math, S4, T4, A4, B4)
sw$S_diff
#> math algebra geometry calculus equations computation
#> 0.003158583 0.003242220 0.001271607 0.031652155 0.003074379 0.016247332
#> numbers addition
#> 0.035000510 -0.010817083
sw$T_diff
#> poetry art dance literature novel
#> -0.0265718087 0.0054876842 -0.0523231481 -0.0117847993 -0.0369267966
#> symphony drama sculpture
#> 0.0224587349 -0.0167662057 0.0003334358
Created on 2022-01-27 by the reprex package (v2.0.1)
Can the
S_diff
andT_diff
components of the output for the method have names? I think each value corresponds to an input term, yes? Would be more useful as named vectors.re: https://github.com/openjournals/joss-reviews/issues/4036