Development and Research of Methods for the Formation of Personalized Bank Offers for an Electronic Queue Terminal

Banks follow all trends and implement Iot technologies, optimize processes, increasing the speed of service. A “smart” feasibility study will optimize the process of lining up a client in a queue, reducing the client's time with the terminal. The purpose of the dissertation is to explore machine learning methods to create a personalized recommendation module for a feasibility study. Improving the quality of customer service in bank branches by increasing the speed of using the feasibility study. Within the framework of this work, a review of existing analogues, analysis and study of the k-nearest neighbors’ method, Support Vector Machines, Random Forests, Naive Bayes, Linear Discriminant Analysis, Logistic Regression, Adaptive Boosting based on feasibility study data. A prototype of the Smart Feasibility Study Web site with a connection to the database has been created. Testing of the website was carried out and the result was obtained according to the assessment of the subjects - an increase in the speed of work. The study consists of 97 pages, including 70 figures, 8 tables, 6 graphs, 24 references. It also includes introduction, 8 chapters and conclusion.