On Predicting Ticket Reopening for Improving Customer Service in 5G Fiber Optic Networks

The paper proposes a data-driven strategy for predicting technical ticket reopening in the context of customer service telecommunications companies providing 5G fiber optic networks. Namely, main aim is to ensure that, between end user and provider, Service Level Agreement terms perceived Quality satisfied. activity has been carried out within framework an extensive joint research initiative focused on Next Generation Networks ELIS Innovation Hub major network provider Italy over years 2018–2021. authors make detailed comparison among performance different approaches classification—ranging from decision trees Artificial Neural Support Vector Machines—and claim that Bayesian classifier most accurate at whether monitored will be reopened or not. Moreover, propose approach dimensionality reduction proves successful increasing computational efficiency, namely by reducing size relevant training dataset two orders magnitude with respect original dataset. Numerical simulations paper, proving proposed can very useful tool providers order identify customers are risk due unsolved issue.