pcut_threshold: Proportional Thresholding (PCut)

In rivolli/utiml: Utilities for Multi-Label Learning

Description

Define the proportion of examples for each label will be positive. The Proportion Cut (PCut) method can be a label-wise or global method that calibrates the threshold(s) from the training data globally or per label.

Usage

pcut_threshold(prediction, ratio, probability = FALSE)

## Default S3 method:
pcut_threshold(prediction, ratio, probability = FALSE)

## S3 method for class 'mlresult'
pcut_threshold(prediction, ratio, probability = FALSE)

Arguments

prediction	A matrix or mlresult.
ratio	A single value between 0 and 1 or a list with ratio values contained one value per label.
probability	A logical value. If TRUE the predicted values are the score between 0 and 1, otherwise the values are bipartition 0 or 1. (Default: FALSE)

Value

A mlresult object.

Methods (by class)

• default: Proportional Thresholding (PCut) method for matrix

• mlresult: Proportional Thresholding (PCut) for mlresult

References

Al-Otaibi, R., Flach, P., & Kull, M. (2014). Multi-label Classification: A Comparative Study on Threshold Selection Methods. In First International Workshop on Learning over Multiple Contexts (LMCE) at ECML-PKDD 2014.

Largeron, C., Moulin, C., & Gery, M. (2012). MCut: A Thresholding Strategy for Multi-label Classification. In 11th International Symposium, IDA 2012 (pp. 172-183).

See Also

Other threshold: fixed_threshold(), lcard_threshold(), mcut_threshold(), rcut_threshold(), scut_threshold(), subset_correction()

Examples

prediction <- matrix(runif(16), ncol = 4)
pcut_threshold(prediction, .45)

rivolli/utiml documentation built on June 1, 2021, 11:48 p.m.