Probabilistic models are models that learn the probability distribution of the underlying data, i.e. automatically detect and store all possible statistical relationships between any number of attributes. The model is based only on the available data itself. Such a model cannot know with certainty whether a particular combination of attributes is impossible or certain. With more observations, the model is more confident and in the case of deterministic rules can estimate them with very high probability (<100%). Thus, the synthetic data generated by the probabilistic model may contain several samples that do not satisfy the deterministic rules. In order to always correctly preserve deterministic rules in the synthetic data, the synthetic model needs additional information about this business knowledge.