Learn R Programming

UBL (version 0.0.9)

An Implementation of Re-Sampling Approaches to Utility-Based Learning for Both Classification and Regression Tasks

Description

Provides a set of functions that can be used to obtain better predictive performance on cost-sensitive and cost/benefits tasks (for both regression and classification). This includes re-sampling approaches that modify the original data set biasing it towards the user preferences.

Copy Link

Version

Install

install.packages('UBL')

Monthly Downloads

663

Version

0.0.9

License

GPL (>= 2)

Issues

10

Pull Requests

1

Stars

32

Forks

11

Maintainer

Paula Branco

Last Published

October 7th, 2023

Functions in UBL (0.0.9)

UBL-package

UBL: Utility-Based Learning
WERCSClassif

WEighted Relevance-based Combination Strategy (WERCS) algorithm for imbalanced classification problems
SMOGNRegress

SMOGN algorithm for imbalanced regression problems
OSSClassif

One-sided selection strategy for handling multiclass imbalanced problems.
NCLClassif

Neighborhood Cleaning Rule (NCL) algorithm for multiclass imbalanced problems
ReBaggRegress

REBaggRegress: RE(sampled) BAG(ging), an ensemble method for dealing with imbalanced regression problems.
TomekClassif

Tomek links for imbalanced classification problems
WERCSRegress

WEighted Relevance-based Combination Strategy (WERCS) algorithm for imbalanced regression problems
UtilInterpol

Utility surface obtained through methods for spatial interpolation of points.
SMOGNClassif

SMOGN algorithm for imbalanced classification problems
UtilOptimClassif

Optimization of predictions utility, cost or benefit for classification problems.
predict,BagModel-method

Predicting on new data with a BagModel model
neighbours

Computation of nearest neighbours using a selected distance function.
phi

Relevance function.
phi.control

Estimation of parameters used for obtaining the relevance function.
RandOverRegress

Random over-sampling for imbalanced regression problems
RandOverClassif

Random over-sampling for imbalanced classification problems
UtilOptimRegress

Optimization of predictions utility, cost or benefit for regression problems.
GaussNoiseClassif

Introduction of Gaussian Noise for the generation of synthetic examples to handle imbalanced multiclass problems.
GaussNoiseRegress

Introduction of Gaussian Noise for the generation of synthetic examples to handle imbalanced regression problems
RandUnderClassif

Random under-sampling for imbalanced classification problems
SmoteClassif

SMOTE algorithm for unbalanced classification problems
RandUnderRegress

Random under-sampling for imbalanced regression problems
SmoteRegress

SMOTE algorithm for imbalanced regression problems
CNNClassif

Condensed Nearest Neighbors strategy for multiclass imbalanced problems
ImbC

Synthetic Imbalanced Data Set for a Multi-class Task
ImbR

Synthetic Regression Data Set
BaggingRegress

Standard Bagging ensemble for regression problems.
AdasynClassif

ADASYN algorithm for unbalanced classification problems, both binary and multi-class.
BagModel-class

Class "BagModel"
distances

Distance matrix between all data set examples according to a selected distance metric.
EvalClassifMetrics

Utility metrics for assessing the performance of utility-based classification tasks.
EvalRegressMetrics

Utility metrics for assessing the performance of utility-based regression tasks.
ENNClassif

Edited Nearest Neighbor for multiclass imbalanced problems