Amazon Textract and artficial intelegence system at banking document managment system

Abstract

Electronic document management systems have a great prospect of use in the banking sector, all information stored in electronic document management systems requires further analysis and processing, this involves the use of a machine learning service to build a semantic search result, which implies the presence of a search service with the thinking of artificial intelligence and the ability provide links to clearly reasoned answers. Such a service that satisfies the needs of semantic search is the Amazon Kendra service, the question of using such a service is more relevant than ever for the construction of modern banking products. Under such conditions, an important area of research is the assessment of the efficiency of Amazon Kendra in the banking sector, which necessitates the development of a conceptual model for assessing the efficiency of banks for making management decisions aimed at improving the efficiency of individual banks and the banking system as a whole. Objectives: The purpose of this work is to improve the work of electronic document flow in the banking sector using Amazon Kendra and Amazon Textract to design an innovative banking product and develop the banking sector of Ukraine. Methods/Approach Scientific research methods – both comparative and analytical – is used in the process of drawing up of this article. Results: A semantic search system based on the bank's electronic document flow system was designed