Empirical Theories Discovery based on the Measurement Theory

The purpose of this work is to analise the cognitive process of the domain theories as it described in the Measurement Theorry and to develop a computational Machine Learning approach, which realise it. As a result the Relational Data Mining approach, proposed by the authors in the preceding books, is developed. We present this approach, as implementation of the cognitive process as it perceived in the Measurement Thoery. In the first part of the paper we analise the cognitive process. In the second we persent the theory and method of the logically most powerful empirical theory discovery. It is based on the notion of “law-like” rules, which satisfy all the properties of laws of nature: generality, simplicity, maximum refutability and minimum number of parameters. This notion is defined for deterministic and probabilistic cases. Based on this method the system “Discovery” is developed. This system was succesfully applied for solving of many practical tasks.