The logical structure of physicochemical data: Problems of numerical data standardization and exchange

摘要

The creation of voluminous databases on the physicochemical properties of substances requires a solution to the general problem of the variability of the logical structure of data. Logical structure variations, which manifest themselves by the differences in the volume, type, and content of data, the appearance of new characteristics, and changes in the rules for the identification of substances, can be caused by many reasons related to both the diversity of types of compounds and the necessity of the integration of data from various sources. The concept of semistructured data is considered; such data are optimum for the transfer of information about the properties of substances. The general principles for constructing the structure of data are formulated. These principles allow tuning to an arbitrary region of objects with the special features of objects (substances or atom-molecular structures) characteristic of it. The suggested concept is compared with the language developed recently within the framework of the IUPAC project for the standardization of exchange and dissemination of thermodynamic data.