Using unlabeled data to help supervised learning has become an increasingly attractive methodology and proven to be effective in many applications. This paper applies semi-supervi...
Traditional supervised classification algorithms require a large number of labelled examples to perform accurately. Semi-supervised classification algorithms attempt to overcome t...