The large number of studies published in the last ten years on the problem of customers\nmigrating from one telecommunications service provider to another competing provider\nproves that this problem has become a major concern for this industry and beyond. The\npurpose of this paper is to detect which variables from the multitude presented in the data\nset for postpaid clients, represents an important driver in the problem of migrating customers\nto another Romanian mobile telecommunications company. To enable us to understand and\nsolve the problem of churn in telecommunications, we need tools that can interpret the\nresults. Thus, we use a Balanced Random Forest for the churn model and three feature\nselection tools: Permutation Importance, Partial Dependence Plot and SHAP. Applying them\nto the churn model, we classify the predictive indicators according to their importance, their\npredictive power and the distribution of the impact that each characteristic has in the model.\nAccording to the Permutation Importance, the drivers regarding churn issue are: the number\nof months since the last offer was changed from the account, the number of minutes\nconsumed outside the company, the value of the invoice, the age of the customer and his time\nat this telecommunications operator. Partially Dependence Plot determinates the churn risk\nareas faced by the Romanian telecommunications company for each of the indicators listed,\nsuch as: clients with younger ages or with outdated offers (unchanged for almost two years).\nSHAP also shows that many months since the last offer, a significant percentage of minutes\nreceived from competing networks or a small age in the network, increases the estimated\nchurn per customer.
Loading....