Title:
Modeling the Customer Experience Using Machine Learning for Optimizing the Performance of Telecommunications Networks: Case of Mobile Networks

Authors:
Alphonse Binele Abana, Patrick Dany Bavoua Kenfack, Paul Salomon Ngohe Ekam, Emmanuel Tonye, Lionel NYOBE MAKANI, Cameroon

Abstract:

Customer experience is a major issue for telecommunications companies seeking to offer quality and personalized services in an environment where customer expectations are constantly evolving. Measuring this experience is a complex challenge, particularly with regard to the impact of network performance on customer usage and, consequently, on their satisfaction. In this context, an in-depth analysis on the relationships between network performance indicators and customer experience was carried out in the network of the telecommunications operator Orange Cameroon (OCM). A method based on machine learning and deep learning was used to predict the behavior of network performance indicators. QoE value thresholds and network performance indicators have been defined for different levels of customer satisfaction. A machine learning model made it possible to obtain precise predictions of customer satisfaction. The results obtained showed that customer satisfaction can be reliably predicted from network performance indicators and QoE. This approach offers a better understanding of the customer experience and allows telecommunications operators to act proactively to improve user satisfaction across all actions related to their use. The application of the results of this study presents a significant impact at different levels. At the national level, telecommunications operators can optimize their services and react more quickly to network incidents, thus improving customer satisfaction and loyalty. This strengthens the competitiveness of telecommunications companies by providing a better customer experience. Globally, a better understanding of customer experience is driving the evolution of the telecommunications industry as a whole

DOI: http://dx.doi.org/10.51505/ijaemr.2024.9511