FMM: Compute F-measure Multilabel
In gecko515/HEMDAG: Hierarchical Ensemble Methods for Directed Acyclic Graphs

Description Usage Arguments Details Value Examples

Function to compute the best hierarchical F-score either one-shot or averaged across folds

1
2
3

compute.Fmeasure.multilabel(target, predicted, n.round = 3,
  f.criterion = "F", verbose = TRUE, b.per.example = FALSE,
  folds = NULL, seed = NULL)

`target`	matrix with the target multilabel: rows correspond to examples and columns to classes. target[i,j]=1 if example i belongs to class j, target[i,j]=0 otherwise.
`predicted`	a numeric matrix with predicted values (scores): rows correspond to examples and columns to classes.
`n.round`	number of rounding digits to be applied to predicted (`default=3`).
`f.criterion`	character. Type of F-measure to be used to select the best F-score. There are two possibilities: `F` (def.) corresponds to the harmonic mean between the average precision and recall; `avF` corresponds to the per-example F-score averaged across all the examples;
`verbose`	boolean. If `TRUE` (def.) the number of iterations are printed on stdout.
`b.per.example`	boolean. `TRUE`: results are returned for each example; `FALSE`: only the average results are returned;
`folds`	number of folds on which computing the AUROC. If `folds=NULL` (`def.`), the AUROC is computed one-shot, otherwise the AUROC is computed averaged across folds.
`seed`	initialization seed for the random generator to create folds. Set `seed` only if `folds`\neq`NULL`. If `seed=NULL` and `folds`\neq`NULL`, the AUROC averaged across folds is computed without seed initialization.

Names of rows and columns of target and predicted matrix must be provided in the same order, otherwise a stop message is returned.

Two different outputs respect to the input parameter b.per.example:

b.per.example==FALSE: a list with a single element average. A named vector with 7 elements relative to the best result in terms of the F.measure: Precision (P), Recall (R), Specificity (S), F.measure (F), av.F.measure (av.F), Accuracy (A) and the best selected Threshold (T). F is the F-measure computed as the harmonic mean between the average precision and recall; av.F is the F-measure computed as the average across examples and T is the best selected threshold;
b.per.example==FALSE: a list with two elements:
1. average: a named vector with with 7 elements relative to the best result in terms of the F.measure: Precision (P), Recall (R), Specificity (S), F.measure (F), av.F.measure (av.F), Accuracy (A) and the best selected Threshold (T);
2. per.example: a named matrix with the Precision (P), Recall (R), Specificity (S), Accuracy (A), F-measure (F), av.F-measure (av.F) and the best selected Threshold (T) for each example. Row names correspond to examples, column names correspond respectively to Precision (P), Recall (R), Specificity (S), Accuracy (A), F-measure (F), av.F-measure (av.F) and the best selected Threshold (T);

data(graph);
data(labels);
data(scores);
root <- root.node(g);
L <- L[,-which(colnames(L)==root)];
S <- S[,-which(colnames(S)==root)];
FMM <- compute.Fmeasure.multilabel(L, S, n.round=3, f.criterion="F", verbose=TRUE, 
b.per.example=TRUE, folds=5, seed=23);