One-Sided Selection

step_oss() creates a specification of a recipe step that removes majority class observations using One-Sided Selection, combining Condensed Nearest Neighbors and Tomek's links.

Usage

step_oss(
  recipe,
  ...,
  role = NA,
  trained = FALSE,
  column = NULL,
  distance = "euclidean",
  skip = TRUE,
  seed = sample.int(10^5, 1),
  distance_with = recipes::all_predictors(),
  id = rand_id("oss")
)

Arguments

recipe

A recipe object. The step will be added to the sequence of operations for this recipe.

...

One or more selector functions to choose which variable is used to sample the data. See recipes::selections for more details. The selection should result in single factor variable. For the tidy method, these are not currently used.

role

Not used by this step since no new variables are created.

trained

A logical to indicate if the quantities for preprocessing have been estimated.

column

A character string of the variable name that will be populated (eventually) by the ... selectors.

distance

A character string specifying the distance metric used for nearest neighbor calculations, defaulting to "euclidean". The available metrics fall into three groups.

"euclidean", "cosine", and "mahalanobis" use approximate nearest neighbors via the RANN package and scale well to large datasets.

"squared_chord", "matusita", "hellinger", and "bhattacharyya" are probability-divergence measures that treat each row as a distribution over the predictors, so they require non-negative values. "hellinger" and "bhattacharyya" further require each row to sum to 1. All four also use the RANN package and scale well to large datasets.

"manhattan", "chebyshev", "canberra", "soergel", "lorentzian", "jeffreys", "topsoe", "jensen-shannon", "jensen_difference", "taneja", and "kumar-johnson" compute an exact all-pairs distance matrix. This takes time and memory proportional to the square of the number of observations in a class, so these are best suited to smaller datasets. Everything from "canberra" onwards is a probability divergence requiring non-negative values, is provided by the philentropy package (which must be installed separately), and in the case of "jeffreys", "taneja", and "kumar-johnson" requires strictly positive values, since those divide by individual predictor values.

The probability divergences are meaningful for compositional predictors such as proportions or counts normalized per observation, and are generally not appropriate for standardized predictors.

skip

A logical. Should the step be skipped when the recipe is baked by bake()? While all operations are baked when prep() is run, some operations may not be able to be conducted on new data (e.g. processing the outcome variable(s)). Care should be taken when using skip = TRUE as it may affect the computations for subsequent operations.

seed

An integer that will be used as the seed when applied.

distance_with

A call to a selector function to choose which variables are used for distance calculations. Defaults to recipes::all_predictors(). The variable selected by ... is always excluded from the distance calculations.

id

A character string that is unique to this step to identify it.

Value

An updated version of recipe with the new step added to the sequence of existing steps (if any). For the tidy method, a tibble with columns terms which is the variable used to sample.

Details

One-Sided Selection (OSS) is an under-sampling method that combines two cleaning techniques. It first applies Condensed Nearest Neighbors (CNN) to reduce the majority classes to a consistent subset that correctly classifies the data using a 1-nearest-neighbor rule, discarding redundant interior observations. It then applies Tomek's links to the remaining observations, removing the majority class observations that form Tomek links with minority class observations, cleaning the decision boundary.

The smallest class is treated as the minority class and is always kept. Because the CNN step relies on a random seed observation and a random scan order, results depend on the random seed.

With more than two classes, the Tomek's links step removes both members of a majority-majority link, not only links between a majority and the minority class. The binary case, the primary intended use, is unaffected.

All variables selected by distance_with must be numeric with no missing data.

All columns in the data are sampled and returned by recipes::juice() and recipes::bake().

When used in modeling, users should strongly consider using the option skip = TRUE so that the extra sampling is not conducted outside of the training set.

Tidying

When you tidy() this step, a tibble is returned with columns terms and id:

terms

character, the selectors or variables selected

id

character, id of this step

Case weights

The underlying operation does not allow for case weights. Supplying data with a case weights column to this step results in an error.

References

Kubat, M., & Matwin, S. (1997). Addressing the curse of imbalanced training sets: one-sided selection. In ICML (Vol. 97, pp. 179-186).

See also

oss() for direct implementation

Other Steps for under-sampling: step_cluster_centroids(), step_cnn(), step_downsample(), step_enn(), step_instance_hardness(), step_ncl(), step_nearmiss(), step_tomek()

Examples

library(recipes)
library(modeldata)
data(hpc_data)

hpc_data0 <- hpc_data |>
  select(-protocol, -day)

orig <- count(hpc_data0, class, name = "orig")
orig
#> # A tibble: 4 × 2
#>   class  orig
#>   <fct> <int>
#> 1 VF     2211
#> 2 F      1347
#> 3 M       514
#> 4 L       259

up_rec <- recipe(class ~ ., data = hpc_data0) |>
  step_oss(class) |>
  prep()

training <- up_rec |>
  bake(new_data = NULL) |>
  count(class, name = "training")
training
#> # A tibble: 4 × 2
#>   class training
#>   <fct>    <int>
#> 1 VF         397
#> 2 F          562
#> 3 M          270
#> 4 L          259

# Since `skip` defaults to TRUE, baking the step has no effect
baked <- up_rec |>
  bake(new_data = hpc_data0) |>
  count(class, name = "baked")
baked
#> # A tibble: 4 × 2
#>   class baked
#>   <fct> <int>
#> 1 VF     2211
#> 2 F      1347
#> 3 M       514
#> 4 L       259

orig |>
  left_join(training, by = "class") |>
  left_join(baked, by = "class")
#> # A tibble: 4 × 4
#>   class  orig training baked
#>   <fct> <int>    <int> <int>
#> 1 VF     2211      397  2211
#> 2 F      1347      562  1347
#> 3 M       514      270   514
#> 4 L       259      259   259

library(ggplot2)

ggplot(circle_example, aes(x, y, color = class)) +
  geom_point() +
  labs(title = "Without OSS") +
  xlim(c(1, 15)) +
  ylim(c(1, 15))


recipe(class ~ x + y, data = circle_example) |>
  step_oss(class) |>
  prep() |>
  bake(new_data = NULL) |>
  ggplot(aes(x, y, color = class)) +
  geom_point() +
  labs(title = "With OSS") +
  xlim(c(1, 15)) +
  ylim(c(1, 15))