Evaluation of Customer Orientation of Russian Companies Using Machine Learning Methods

In this paper, the researchers' attention was focused on how the strategy of customer orientation is reflected in the internal documents of companies and materials posted on the websites. The authors used a formalized method of studying text information-content analysis. The results of the pilot study indicate that for a number of reasons, the client-oriented approach is not a business ideology for many large companies; most of the sample studied is characterized by a formally stated interest in the needs of customers. In order to ensure the possibility of carrying out large - scale regular research, the concept of building a software package for the analysis of customer orientation companies using machine learning methods – SP ACOC has been developed. To implement this concept, the quality metric of classification algorithms was determined; several algorithms were trained; a program was developed and tested, which determines the degree of customer orientation of the company according to its documents according to the chosen algorithm Automation of the process of evaluation of customer orientation companies will reduce its complexity, expand the range of objects studied, improve the accuracy and objectivity of the results.