“…More analytically, we evaluated the performances of self-training and Yet Another Two Stage Idea (YATSI) approaches with the most classic classification methods. Self-training and YATSI constitute two of the most efficient and frequently utilized semisupervised algorithms which have been successfully used in a variety of real-world applications (Catal & Diri, 2009;Driessens, Reutemann, Pfahringer, & Leschi, 2006;Levatic, Dzeroski, Supek, & Smuc, 2013;Roli & Marcialis 2006;Rosenberg, Hebert, & Schneiderman, 2005;Sigdel et al, 2014) providing some promising classification results. Our preliminary numerical experiments illustrate that the classification accuracy can be significantly improved, utilizing a few labeled and many unlabeled data for developing reliable prediction models.…”