EMMIXSSL: Semi-Supervised Learning via Gaussian Mixture Model
The algorithm of semi-supervised learning for a partially classified sample via Gaussian mixture model with the missing-label mechanism is designed for a fitting g-component Gaussian mixture model via maximum likelihood (ML). The classifier is proposed to treat the labels of the unclassified features as missing-data and to introduce a framework for their missing as in the pioneering work of Rubin (1976) for missing in incomplete data analysis. It suggests that the missingness of the labels of the features can be modelled by representing the probability of a missing-label for a feature via the logistic model depending on the entropy of the feature or an appropriate proxy for it.
||R (≥ 3.1.0), mvtnorm, stats
||Ziyang Lyu, Daniel Ahfock, Geoffrey J. McLachlan
||Ziyang Lyu <ziyang.lyu at unsw.edu.au>
Please use the canonical form
to link to this page.