Development of Recommendation System in e-Commerce using Emotional Analysis and Machine Learning Methods

The article identifies the theoretical aspects of e-commerce as an economic category, the problems of the current state of the e-commerce market related to the COVID-19 pandemic, consumer distrust and less functionality in the choice of clothing units. The aim of the article is the generalization, development and implementation of an expert system that will increase the functionality and efficiency of e- business. Research methods are mathematical apparatus for solving classification problems, regression, discrete methods, artificial neural networks, decision trees and other machine learning algorithms for solving classification problems, statistical methods and tools for data processing and analysis, data visualization tools. The problems of developing an expert system to increase the functionality of e-commerce were solved using R and Python programming languages, data processing and analysis. Practical significance of the research is to identify and use an effective algorithm for emotional analysis of the text, improve the method of collaborative filtering and develop an information system to provide recommendations to customers. As a result, the functionality and efficiency of e-commerce will increase. The obtained research results can be used as a methodological reference by software developers during the design and development of recommendation systems.