Implementations of Semi-Supervised Learning Approaches for Classification

Documentation for package ‘RSSL’ version 0.9.6

Help Pages

-- A --

add_missinglabels_mar	Throw out labels at random
adjacency_knn	Calculate knn adjacency matrix

-- B --

Classifier used for enabling shared documenting of parameters

-- C --

c.CrossValidation	Merge result of cross-validation runs on single datasets into a the same object
clapply	Use mclapply conditional on not being in RStudio
cov_ml	Biased (maximum likelihood) estimate of the covariance matrix
CrossValidationSSL	Cross-validation in semi-supervised setting
CrossValidationSSL.list	Cross-validation in semi-supervised setting
CrossValidationSSL.matrix	Cross-validation in semi-supervised setting

-- D --

decisionvalues	Decision values returned by a classifier for a set of objects
decisionvalues-method	Decision values returned by a classifier for a set of objects
decisionvalues-method	Predict using RSSL classifier
df_to_matrices	Convert data.frame with missing labels to matrices
diabetes	diabetes data for unit testing

-- E --

EMLeastSquaresClassifier	An Expectation Maximization like approach to Semi-Supervised Least Squares Classification
EMLinearDiscriminantClassifier	Semi-Supervised Linear Discriminant Analysis using Expectation Maximization
EMNearestMeanClassifier	Semi-Supervised Nearest Mean Classifier using Expectation Maximization
EntropyRegularizedLogisticRegression	Entropy Regularized Logistic Regression

-- F --

Find a violated label

-- G --

gaussian_kernel	calculated the gaussian kernel matrix
generate2ClassGaussian	Generate data from 2 Gaussian distributed classes
generateABA	Generate data from 2 alternating classes
generateCrescentMoon	Generate Crescent Moon dataset
generateFourClusters	Generate Four Clusters dataset
generateParallelPlanes	Generate Parallel planes
generateSlicedCookie	Generate Sliced Cookie dataset
generateSpirals	Generate Intersecting Spirals
generateTwoCircles	Generate data from 2 circles
geom_classifier	Plot RSSL classifier boundary (deprecated)
geom_linearclassifier	Plot linear RSSL classifier boundary
GRFClassifier	Label propagation using Gaussian Random Fields and Harmonic functions

-- H --

Direct R Translation of Xiaojin Zhu's Matlab code to determine harmonic solution

-- I --

ICLeastSquaresClassifier	Implicitly Constrained Least Squares Classifier
ICLinearDiscriminantClassifier	Implicitly Constrained Semi-supervised Linear Discriminant Classifier

-- K --

KernelICLeastSquaresClassifier	Kernelized Implicitly Constrained Least Squares Classification
KernelLeastSquaresClassifier	Kernelized Least Squares Classifier

-- L --

LaplacianKernelLeastSquaresClassifier	Laplacian Regularized Least Squares Classifier
LaplacianSVM	Laplacian SVM classifier
LearningCurveSSL	Compute Semi-Supervised Learning Curve
LearningCurveSSL.matrix	Compute Semi-Supervised Learning Curve
LeastSquaresClassifier	Least Squares Classifier
LinearDiscriminantClassifier	Linear Discriminant Classifier
LinearSVM	Linear SVM Classifier
LinearSVM-class	LinearSVM Class
LinearTSVM	Linear CCCP Transductive SVM classifier
line_coefficients	Loss of a classifier or regression function
line_coefficients-method	Loss of a classifier or regression function
localDescent	Local descent
LogisticLossClassifier	Logistic Loss Classifier
LogisticLossClassifier-class	LogisticLossClassifier
LogisticRegression	(Regularized) Logistic Regression implementation
LogisticRegressionFast	Logistic Regression implementation that uses R's glm
logsumexp	Numerically more stable way to calculate log sum exp
loss	Loss of a classifier or regression function
loss-method	Loss of a classifier or regression function
losslogsum	LogsumLoss of a classifier or regression function
losslogsum-method	LogsumLoss of a classifier or regression function
losspart	Loss of a classifier or regression function evaluated on partial labels
losspart-method	Loss of a classifier or regression function evaluated on partial labels

-- M --

MajorityClassClassifier	Majority Class Classifier
MCLinearDiscriminantClassifier	Moment Constrained Semi-supervised Linear Discriminant Analysis.
MCNearestMeanClassifier	Moment Constrained Semi-supervised Nearest Mean Classifier
MCPLDA	Maximum Contrastive Pessimistic Likelihood Estimation for Linear Discriminant Analysis
measure_accuracy	Performance measures used in classifier evaluation
measure_error	Performance measures used in classifier evaluation
measure_losslab	Performance measures used in classifier evaluation
measure_losstest	Performance measures used in classifier evaluation
measure_losstrain	Performance measures used in classifier evaluation
minimaxlda	Implements weighted likelihood estimation for LDA
missing_labels	Access the true labels for the objects with missing labels when they are stored as an attribute in a data frame

-- N --

Nearest Mean Classifier

-- P --

plot.CrossValidation	Plot CrossValidation object
plot.LearningCurve	Plot LearningCurve object
posterior	Class Posteriors of a classifier
posterior-method	Class Posteriors of a classifier
predict-method	Predict for matrix scaling inspired by stdize from the PLS package
predict-method	Predict using RSSL classifier
PreProcessing	Preprocess the input to a classification function
PreProcessingPredict	Preprocess the input for a new set of test objects for classifier
print.CrossValidation	Print CrossValidation object
print.LearningCurve	Print LearningCurve object
projection_simplex	project an n-dim vector y to the simplex Dn

-- Q --

Quadratic Discriminant Classifier

-- R --

responsibilities	Responsibilities assigned to the unlabeled objects
responsibilities-method	Predict using RSSL classifier
RSSL	R Semi-Supervised Learning Package
rssl-formatting	Show RSSL classifier
rssl-predict	Predict using RSSL classifier

-- S --

S4VM	Safe Semi-supervised Support Vector Machine (S4VM)
S4VM-class	LinearSVM Class
sample_k_per_level	Sample k indices per levels from a factor
scaleMatrix	Matrix centering and scaling
SelfLearning	Self-Learning approach to Semi-supervised Learning
show-method	Show RSSL classifier
solve_svm	SVM solve.QP implementation
split_dataset_ssl	Create Train, Test and Unlabeled Set
split_random	Randomly split dataset in multiple parts
SSLDataFrameToMatrices	Convert data.frame to matrices for semi-supervised learners
stat_classifier	Plot RSSL classifier boundaries
stderror	Calculate the standard error of the mean from a vector of numbers
summary.CrossValidation	Summary of Crossvalidation results
svdinv	Inverse of a matrix using the singular value decomposition
svdinvsqrtm	Taking the inverse of the square root of the matrix using the singular value decomposition
svdsqrtm	Taking the square root of a matrix using the singular value decomposition
SVM	SVM Classifier
svmlin	svmlin implementation by Sindhwani & Keerthi (2006)
svmlin_example	Test data from the svmlin implementation
svmproblem	Train SVM

-- T --

testdata	Example semi-supervised problem
threshold	Refine the prediction to satisfy the balance constraint
true_labels	Access the true labels when they are stored as an attribute in a data frame
TSVM	Transductive SVM classifier using the convex concave procedure

-- U --

USMLeastSquaresClassifier	Updated Second Moment Least Squares Classifier
USMLeastSquaresClassifier-class	USMLeastSquaresClassifier

-- W --

wdbc	wdbc data for unit testing
WellSVM	WellSVM for Semi-supervised Learning
wellsvm_direct	wellsvm implements the wellsvm algorithm as shown in [1].
WellSVM_SSL	Convex relaxation of S3VM by label generation
WellSVM_supervised	A degenerated version of WellSVM where the labels are complete, that is, supervised learning
wlda	Implements weighted likelihood estimation for LDA
wlda_error	Measures the expected error of the LDA model defined by m, p, and iW on the data set a, where weights w are potentially taken into account
wlda_loglik	Measures the expected log-likelihood of the LDA model defined by m, p, and iW on the data set a, where weights w are potentially taken into account