EMNearestMeanClassifier {RSSL}

Semi-Supervised Nearest Mean Classifier using Expectation Maximization

Description

Expectation Maximization applied to the nearest mean classifier assuming Gaussian classes with a spherical covariance matrix.

Usage

EMNearestMeanClassifier(X, y, X_u, method = "EM", scale = FALSE,
  eps = 1e-04)

Arguments

X

matrix; Design matrix for labeled data

y

factor or integer vector; Label vector

X_u

matrix; Design matrix for unlabeled data

method

character; Currently only "EM"

scale

Should the features be normalized? (default: FALSE)

eps

Stopping criterion for the maximinimization

Details

Starting from the supervised solution, uses the Expectation Maximization algorithm (see Dempster et al. (1977)) to iteratively update the means and shared covariance of the classes (Maximization step) and updates the responsibilities for the unlabeled objects (Expectation step).

References

Dempster, A., Laird, N. & Rubin, D., 1977. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society. Series B, 39(1), pp.1-38.