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.

Downloads: