Development of Methodology for Analyzing the Legality of Financial Resources on the Internet Based on Machine Learning Methods

The final qualification work is aimed at developing a methodology for determining the legality of Internet resources engaged in financial activities, based on the identified criteria and classifications, with the introduction of machine learning. In this paper, we plan to study the methods and methods of machine learning that are suitable for working with text data. The tasks of last year's interdisciplinary course work on the topic “Development of methods for automated identification of illegal financial Internet resources” were to read identifiers, obtain information about the organization and search for the collected data on the resource pages. However, not all text data was detected by the program, which affected the accuracy of the results. Therefore, to improve the accuracy, it was decided to implement machine learning.