Structural Model of an Innovative Information System for Prevention and Detection of Financial Customs Violations

The article describes the design of a system for detecting financial customs violations using the methods of artificial intelligence. The information is collected automatically from heterogeneous databases of different institutions and stored in a data center. At the output of the system, hypotheses are generated, indicated with the respective probabilities concerning past, present or future customs violations. The result of the work, the intervention of the operators and the actions of the customs officers are recorded and stored according to the block-chain technology. The system is designed in such a way as to control the actions of the human operator, which makes any external intervention very difficult or impossible. The efficiency of knowledge extraction to generate hypotheses is guaranteed by the use of combination of various AI methods.