getSetIntersections.RdCalculate sizes of unions and intersections between each pair of sets
getSetIntersections(df, setNames, idName)
| df | A data frame indicating set membership |
|---|---|
| setNames | A character vector of set names |
| idName | A string specifying name of ID variable for each item |
A data frame with variables:
set1 and set2 pair of sets
Ninter and Nunion number of elements in
the intersection and union of the pair of sets
N1 and N2 number of items in each set
prop1 and prop2 proportion of items in
each set that are also members of the other set
(prop1 = Ninter / N1)
prop proportion of items from either set that
are members of both sets (prop = Ninter / Nunion)
.pred predictions assuming marginal independence of sets
.error difference between predictions and observations
.relError error relative to observations
The input data frame should contain a row for each item and a binary variable
for each set indicating the membership of each item. The setNames
input should correspond to the binary indicator columns in the data frame.
# Define set names data("movieSets") setNames <- colnames(movieSets[,-c(1:8)]) # Calculate set sizes getSetIntersections(movieSets, setNames, "movieId")#> # A tibble: 361 x 20 #> set1 set2 Ntotal Ninter Nunion N1 N2 prop prop1 prop2 #> <fct> <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> #> 1 Acti~ Acti~ 27278 3520 3520 3520 3520 1 1 1 #> 2 Acti~ Adve~ 27278 972 4877 3520 2329 0.199 0.276 0.417 #> 3 Acti~ Anim~ 27278 198 4349 3520 1027 0.0455 0.0562 0.193 #> 4 Acti~ Chil~ 27278 102 4557 3520 1139 0.0224 0.0290 0.0896 #> 5 Acti~ Come~ 27278 719 11175 3520 8374 0.0643 0.204 0.0859 #> 6 Acti~ Crime 27278 767 5692 3520 2939 0.135 0.218 0.261 #> 7 Acti~ Docu~ 27278 11 5980 3520 2471 0.00184 0.00312 0.00445 #> 8 Acti~ Drama 27278 1203 15661 3520 13344 0.0768 0.342 0.0902 #> 9 Acti~ Fant~ 27278 325 4607 3520 1412 0.0705 0.0923 0.230 #> 10 Acti~ Film~ 27278 13 3837 3520 330 0.00339 0.00369 0.0394 #> # ... with 351 more rows, and 10 more variables: Ninter.pred <dbl>, #> # Nunion.pred <dbl>, prop.pred <dbl>, prop1.pred <dbl>, Ninter.error <dbl>, #> # prop.error <dbl>, prop1.error <dbl>, Ninter.relError <dbl>, #> # prop.relError <dbl>, prop1.relError <dbl>