Institute of Mathematics and Informatics Bulgarian Academy of Sciences, Association for the Development of the Information Society
Proceedings of the National Conference on "Education in the Information Society", Plovdiv, May, 2013, 149p-157p
With the development of new technology, information has become easier to access.
Databases consist of numerous examples, reaching new dimensions. New powerful, parallel,
executing fast algorithms are needed. The aim of the presented algorithm is exactly to fulfill
the greed for this trend. A multi-threaded semi-supervised version of the standard EMalgorithm
is presented. It uses two data sources – labeled and unlabeled data. Usually, we
have access to lots of unlabeled examples, but the labeled ones are difficult to reach.
Report published in the Proceedings of the National Conference on "Education in the Information Society", Plovdiv, May, 2013