SemiSupervised: Safe Semi-Supervised Learning Tools

Implements several safe graph-based semi-supervised learning algorithms. The first algorithm is the Semi-Supervised Semi-Parametric Model (S4PM) and the fast Anchor Graph version of this approach. For additional technical details, refer to Culp and Ryan (2013) <http://jmlr.org/papers/v14/culp13a.html>, Ryan and Culp (2015) <http://www.jmlr.org/papers/v16/ryan15a.html> and the package vignette. The underlying fitting routines are executed in C++. All tuning parameter estimation is optimized using K-fold Cross-Validation.

Version: 1.0
Depends: R (≥ 2.10), methods
Suggests: caret, mlbench, kernlab, randomForest, glmnet, spa, e1071
Published: 2018-05-11
Author: Mark Vere Culp
Maintainer: Mark Vere Culp <mvculp at mail.wvu.edu>
License: GPL-2
NeedsCompilation: yes
CRAN checks: SemiSupervised results

Downloads:

Reference manual: SemiSupervised.pdf
Vignettes: A Short Introduction to the SemiSupervised Package
Package source: SemiSupervised_1.0.tar.gz
Windows binaries: r-devel: SemiSupervised_1.0.zip, r-release: SemiSupervised_1.0.zip, r-oldrel: SemiSupervised_1.0.zip
OS X binaries: r-release: SemiSupervised_1.0.tgz, r-oldrel: SemiSupervised_1.0.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=SemiSupervised to link to this page.