Semantic Analysis Models and Methods of the Text Information in the Building Normative Base

The research is aimed at overcoming the uncertainties associated with the reform of legislation regarding the technical regulation of the building industry. The main goal of the paper is to systematize the normative base of building of Ukraine in the conditions of adaptation of normative documents to the requirements of the European Union. In order to achieve this goal, an analysis of the peculiarities of the information resource of the normative base was conducted and the nature of the uncertainty contained in the normative documentation of the industry was investigated. The status of modern electronic databases of normative documentation in the building and building materials industry of Ukraine is described. The scheme for identifying different documents on the same issue is proposed. The possibility of applying models and methods of fuzzy mathematics to formalization of texts of building norms and rules and expression of their semantics in the internal language of the Semantic Text Information Analysis System is shown. The practical value of the paper is seen in the improvement of the organizational and technical system "Technical Regulation in Building" by minimizing the number of regulations on the same issue and achieving consistency between normative. The paper's scientific novelty of is to apply the model of artificial intelligence to the solution of the problem of semantic analysis of documents contained in electronic bases of normative documentation of the system of technical regulation in building. The capability of artificial neural networks of the ImegNet category to solve the problem of semantic analysis text documents of Building Normative Base is proved Further researches will focus on forming the Text Information Semantic Analysis System knowledge base and adapting the Deep Structured Semantic Model to the task of assessing the degree of closeness between a Semantic Request and a Semantic Normative Document. The results can be used to improve other workflow systems.