Abstract:

Customers in the telecom sector have access to a variety of service providers and can actively switch from one operator to another. In this fiercely competitive market, the telecom business has an average annual churn rate of 15 to 25 per cent. Retaining highly profitable consumers is the top business objective for many established operators. Telecom businesses must identify the consumers who are most likely to leave in order to reduce customer turnover. Using the data (features) from the first three months, the business goal is to estimate the churn in the most recent (i.e., ninth) month. Considering the normal consumer behaviour during churn will help with this endeavour. The novel approach frequently employ a Logistic Regression model and Random Forest Model are used to actively capture the business objective of predicting customer churn and also this study used PCA model to find out the most effective features by feature reduction. The accuracies acquired are 83%, 75% respectively. The reasons that the service provider can use to better and completely reduce customer turnover are also provided.

Description:

Machine Learning, PCA, Recall, Recursive Feature Elimination, Telecom Churn