- Notifications
You must be signed in to change notification settings - Fork0
Anonymise factor columns across datasets in a consistent way
License
NotificationsYou must be signed in to change notification settings
dkgaraujo/simplanonym
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
The goal of simplanonym is to easily anonymise individuals that appearin multiple datasets, in a consistent way.
Consistent anonymisation means that:
- each unique individual gets the same single anonymised factor, evenif appearing in more than one dataset;
- the same anonymised factor always refers back to the sameindividual.
You can install simplanonym fromCRAN with:
install.packages("simplanonym")This is a basic example which shows you how to solve a common problem:
library(simplanonym)rand_tbl_1<-vroom::gen_tbl(10,4,col_types="fffd")rand_tbl_2<-vroom::gen_tbl(10,2,col_types="fd")rand_tbl_2$X3<-rand_tbl_1$X3# note:# * rand_tbl_1 and rand_tbl_2 share three column names,# of which X2 is a factor in one but not the other.# * X1 factors do not overlap, but their anonymisation# should still be consistent (ie, different levels should# have their own unique anonymised factors).# * For X3, the anonymised factors should consider the levels# at both `rand_tbl_1$X3` and `rand_tbl_2$X3`.data_list<-list(rand_tbl_1,rand_tbl_2)data_list#> [[1]]#> # A tibble: 10 × 4#> X1 X2 X3 X4#> <fct> <fct> <fct> <dbl>#> 1 different_gorilla thankful_pony scary_mustang 1.53#> 2 different_gorilla own_leopard beautiful_marten 0.204#> 3 purring_donkey witty_tiger straight_deer 1.05#> 4 itchy_badger own_leopard adorable_panther 0.00518#> 5 young_snake calm_finch white_rhinoceros -1.37#> 6 chubby_rabbit public_lizard scary_mustang 1.16#> 7 hot_bunny loose_eagle glamorous_puma -1.83#> 8 gentle_hare deep_goat hissing_puppy -1.09#> 9 brief_weasel green_impala adorable_panther -1.28#> 10 cool_gazelle nutritious_bird beautiful_marten -1.70#>#> [[2]]#> # A tibble: 10 × 3#> X1 X2 X3#> <fct> <dbl> <fct>#> 1 good_chinchilla -0.461 scary_mustang#> 2 defeated_cougar -0.0267 beautiful_marten#> 3 sweet_kangaroo 0.152 straight_deer#> 4 good_chinchilla -0.883 adorable_panther#> 5 good_chinchilla 2.36 white_rhinoceros#> 6 sweet_kangaroo -0.570 scary_mustang#> 7 rotten_bee 0.297 glamorous_puma#> 8 huge_eagle -0.0890 hissing_puppy#> 9 rotten_bee -1.06 adorable_panther#> 10 good_chinchilla -0.308 beautiful_martendata_list|> anonymise(return_original_levels=TRUE)#> Joining, by = c("X1", "X2", "X3")#> Joining, by = c("X1", "X3")#> [[1]]#> # A tibble: 10 × 7#> X1 X2 X3 X1_anon X2_anon X3_anon X4#> <fct> <fct> <fct> <fct> <fct> <fct> <dbl>#> 1 different_gorilla thankful_pony scary_mus… 19 12 12 1.53#> 2 different_gorilla own_leopard beautiful… 19 09 04 0.204#> 3 purring_donkey witty_tiger straight_… 05 02 16 1.05#> 4 itchy_badger own_leopard adorable_… 26 09 19 0.00518#> 5 young_snake calm_finch white_rhi… 21 08 06 -1.37#> 6 chubby_rabbit public_lizard scary_mus… 02 19 12 1.16#> 7 hot_bunny loose_eagle glamorous… 30 11 03 -1.83#> 8 gentle_hare deep_goat hissing_p… 25 21 01 -1.09#> 9 brief_weasel green_impala adorable_… 13 16 19 -1.28#> 10 cool_gazelle nutritious_bird beautiful… 12 07 04 -1.70#>#> [[2]]#> # A tibble: 10 × 5#> X1 X3 X1_anon X3_anon X2#> <fct> <fct> <fct> <fct> <dbl>#> 1 good_chinchilla scary_mustang 04 12 -0.461#> 2 defeated_cougar beautiful_marten 33 04 -0.0267#> 3 sweet_kangaroo straight_deer 23 16 0.152#> 4 good_chinchilla adorable_panther 04 19 -0.883#> 5 good_chinchilla white_rhinoceros 04 06 2.36#> 6 sweet_kangaroo scary_mustang 23 12 -0.570#> 7 rotten_bee glamorous_puma 01 03 0.297#> 8 huge_eagle hissing_puppy 11 01 -0.0890#> 9 rotten_bee adorable_panther 01 19 -1.06#> 10 good_chinchilla beautiful_marten 04 04 -0.308
The hex sticker for the package is based on an icon provided free of charge bywww.flaticon.com.René magritte icons created by Freepik - Flaticon
About
Anonymise factor columns across datasets in a consistent way
Topics
Resources
License
Uh oh!
There was an error while loading.Please reload this page.
Stars
Watchers
Forks
Packages0
No packages published
