SSLfmm: Semi-Supervised Learning with Mixed Missingness in Finite
Mixture Models
Semi-supervised Gaussian finite mixture models for partially labelled data
under complete-case, missing completely at random (MCAR), entropy-dependent missing
at random (MAR), and mixed MCAR/MAR label-missingness formulations. For the mixed
formulation, the source of a missing label may be observed or latent. The package
supports equal and component-specific covariance matrices, model fitting, simulation,
initialization, prediction, classification performance assessment, and entropy-based
diagnostics. A semi-synthetic Blood Transfusion data set is included to illustrate the
applied workflow.
Documentation:
Downloads:
Linking:
Please use the canonical form
https://CRAN.R-project.org/package=SSLfmm
to link to this page.
[aut],
Jinran Wu