We present an experiment of using transformation-based learning for improving translation quality of a rule-based machine translation system by means of post-processing. Transformation rules are learned based on a parallel corpus of machine translation output and a human-corrected version of the output. The experiment resulted in a significant increase in translation quality of 0.8 measured usi...

In this paper, we propose a quality-based use case refinement approach. It consists of a step by step refinement process that combines quality metrics with use case transformation rules. We propose several quality metrics, based on complexity concepts, aimed at measuring the complexity of use cases. Starting from an initial use case, we apply successively a set of transformation rules and measu...

We investigate the effect of power transformation of raw scores on the responsiveness of quality of life survey. The procedure maximizes the paired t-test value on the power transformed data to obtain an optimal power range. The parallel between the Box-Cox transformation is also investigated for the quality of life data.

Problem statement. Educational institutions are actively moving towards digitalization processes. These processes have to be implemented not from scratch, but taking into account the databases, information systems and technologies already applied at informatization stage used. Therefore, issues related concept of digital transformation educational organizations become relevant. The study aimed ...

Solving the convergence issues of Generative Adversarial Networks (GANs) is one most outstanding problems in generative models. In this work, we propose a novel activation function to be used as output generator agent. This based on Smirnov probabilistic transformation and it specifically designed improve quality generated data. sharp contrast previous works, our provides more general approach ...