Movatterモバイル変換

R Semi-Supervised Learningpackage

This R package provides implementations of several semi-supervisedlearning methods, in particular, our own work involving constraint basedsemi-supervised learning.

To cite the package, use either of these two references:

• Krijthe, J. H. (2016). RSSL: R package for Semi-supervised Learning.In B. Kerautret, M. Colom, & P. Monasse (Eds.), ReproducibleResearch in Pattern Recognition. RRPR 2016. Lecture Notes in ComputerScience, vol 10214. (pp. 104–115). Springer International Publishing.https://doi.org/10.1007/978-3-319-56414-2_8. arxiv:https://arxiv.org/abs/1612.07993
• Krijthe, J.H. & Loog, M. (2015). Implicitly ConstrainedSemi-Supervised Least Squares Classification. In E. Fromont, T. de Bie,& M. van Leeuwen, eds. 14th International Symposium on Advances inIntelligent Data Analysis XIV (Lecture Notes in Computer Science Volume9385). Saint Etienne. France, pp. 158-169.

Installation Instructions

This package available on CRAN. The easiest way to install thepackage is to use:

install.packages("RSSL")

To install the latest version of the package using the devtoolspackage:

library(devtools)install_github("jkrijthe/RSSL")

Usage

After installation, load the package as usual:

library(RSSL)

The following code generates a simple dataset, trains a supervisedand two semi-supervised classifiers and evaluates their performance:

library(dplyr,warn.conflicts =FALSE)library(ggplot2,warn.conflicts =FALSE)set.seed(2)df<-generate2ClassGaussian(200,d=2,var =0.2,expected=TRUE)# Randomly remove labelsdf<- df%>%add_missinglabels_mar(Class~.,prob=0.98)# Train classifierg_nm<-NearestMeanClassifier(Class~.,df,prior=matrix(0.5,2))g_self<-SelfLearning(Class~.,df,method=NearestMeanClassifier,prior=matrix(0.5,2))# Plot datasetdf%>%ggplot(aes(x=X1,y=X2,color=Class,size=Class))+geom_point()+coord_equal()+scale_size_manual(values=c("-1"=3,"1"=3),na.value=1)+geom_linearclassifier("Supervised"=g_nm,"Semi-supervised"=g_self)

# Evaluate performance: Squared Loss & Error Ratemean(loss(g_nm,df))mean(loss(g_self,df))mean(predict(g_nm,df)!=df$Class)mean(predict(g_self,df)!=df$Class)

Acknowledgement

Work on this package was supported by Project 23 of the Dutchnational program COMMIT.