Methods for Synthesis of Relational Data Model in Information Systems Reengineering Problems

The purpose of the article is to develop a method for reengineering relational databases, taking into account the presence of implicit interrelated functionally dependent data that affect the structure of the logical model. The following results are obtained: the article proposes an approach to detecting previously unknown functional dependences, based on the analysis of a set of relational database. Classes of relational databases reengineering problems are singled out; the stage of forming the target logical scheme, common to the problems of adaptation and refactoring, is investigated. The sub-task of verification of the relational database logical scheme correspondence to the third normal form within the framework of this stage is considered using the synthesis method; it is shown that its solution involves a number of difficulties, in particular, the need to find the set of functional dependences being performed using the current data instance of some relational database. An approach is proposed for finding a set of functional dependencies from the relational structure data instance. As a direction for further research, one can highlight the implementation of empty values support at the stage of detection of functional dependences, as well as issues of data transfer without loss from the original structure of the database to the target one, obtained as a result of applying the reengineering methods.