SSL: Semi-Supervised Learning

Semi-supervised learning has attracted the attention of machine learning community because of its high accuracy with less annotating effort compared with supervised learning.The question that semi-supervised learning wants to address is: given a relatively small labeled dataset and a large unlabeled dataset, how to design classification algorithms learning from both ? This package is a collection of some classical semi-supervised learning algorithms in the last few decades.

Author
Junxiang Wang
Date of publication
2016-05-14 23:12:09
Maintainer
Junxiang Wang <xianggebenben@163.com>
License
GPL (>= 3)
Version
0.1

View on CRAN

Man pages

sslCoTrain
Co-Training
sslGmmEM
Gaussian Mixture Model with an EM Algorithm
sslLabelProp
Label Propagation
sslLapRLS
Laplacian Regularized Least Squares
sslLDS
Low Density Separation
sslLLGC
Local and Global Consistency
sslMarkovRandomWalks
t-step Markov Random Walks
sslMincut
Mincut
sslRegress
Regression on graphs
sslSelfTrain
Self-Training

Files in this package

SSL
SSL/src
SSL/src/MPLA.cpp
SSL/src/RcppExports.cpp
SSL/src/Floyd.cpp
SSL/NAMESPACE
SSL/R
SSL/R/SSL.R
SSL/R/RcppExports.R
SSL/MD5
SSL/DESCRIPTION
SSL/man
SSL/man/sslGmmEM.Rd
SSL/man/sslCoTrain.Rd
SSL/man/sslLabelProp.Rd
SSL/man/sslMincut.Rd
SSL/man/sslRegress.Rd
SSL/man/sslLLGC.Rd
SSL/man/sslMarkovRandomWalks.Rd
SSL/man/sslSelfTrain.Rd
SSL/man/sslLapRLS.Rd
SSL/man/sslLDS.Rd