Generation of pseudoartificial data sets

We created a generator of pseudo artificial data sets. The generator takes a given data set as input, analyses it and creates a new data set based on learned properties. A key component of the generator is PRBF algorithm, because its structure is particularly suited for example generation. We tested the generator on multiple real data sets and measured its quality with the data sets properties. In the conclusion we propose possible enhancements to generator's abilities.