Search

Digital transformation has affected many industries and has become a mega-trend in the information technology systems. However, the venture capital investment process is still rigid. The use of distributed ledger technology and artificial intelligence in the field of finance and investment requires a detailed framework and a rethinking of the interaction model of market participants. The article shows the DLT based approach with attached of F-BFT consensus to improve the efficiency of the investment process in the modern world.

Modern attribution of art works depends on the expert experience. This implies some subjectivity of his assessments. Recently for making a more informed decision digital image processing techniques have been used. Therefore, the study of the methods for obtaining additional information about the painting, which cannot be obtained using existing methods, is an important task.

The aim of the research is to compare and select software tools for rapid prototyping of face retrieval systems with arbitrary structure (parallel, cascade, combined) that satisfy the requirements of the problem being solved. The relevance of development of modeling tools for face retrieval systems (FaRetSys) was shown, selection of software tools (Simulink + Matlab) was substantiated: methods are previously debugged in Matlab, and then integrated into Simulink blocks. The concept of organizing a library for modeling FaRetSys is proposed. Functions and the methodology of using the library in solving modeling problems are given from point of view the block composition concept. The developed library was used to solve the variational problem of selecting the structure and adjusting the parameters of FaRetSys based on mathematical modeling in Simulink. A synthesized system with a parallel structure confirmed the increased accuracy of FaRetSys by using parallel FaRetSys and making a decision based on a comparison of subsystems operation results.