Optimization of the Management of the Customer Base of a Telecommunications Company Using Artificial Intelligence Methods

Abstract

The aim of the study is to substantiate the use of machine learning and statistical analysis methods, in particular the CHAID (Chi-squared Automatic Interaction Detection) algorithm, to identify key factors influencing customer churn and telecommunications company revenue. The research is directed towards developing effective strategies for managing the customer base and optimizing business processes in the telecommunications industry. The article conducts a comprehensive analysis of the client base of a telecommunications company using the method of decision trees. The six most important factors that have the greatest influence on customers’ decisions to continue or terminate the use of services have been identified: type of contract, type of Internet service, duration of use of services, use of movie streaming services, method of payment and retiree status. The study found that these factors affect not only customer churn but also the company’s revenue from each customer. This highlights the importance of a comprehensive approach to analyzing the customer base, which takes into account both churn risks and the financial aspects of customer interactions. Based on the results obtained, the introduction of an individual approach to clients with different characteristics has been proposed. Such a strategy will allow you to more effectively meet the needs of different segments of the customer base, increase their loyalty and maximize the company’s revenue. The study opens up prospects for further research in the direction of optimizing customer base management, in particular in the development of methods for interpreting machine learning models, improving methods for segmenting the customer base, and developing dynamic models that take into account changes in customer behavior over time.

Keywords

• artificial intelligence
• customer base management
• telecommunications
• customer segmentation
• churn forecasting
• machine learning
• data analysis
• chaid algorithm
• decision trees
• business process optimization
• customer loyalty
• service personalization
• statistical analysis
• model interpretation
• dynamic segmentation models