Intelligent Framework for Detection of Telecom-Customer Churn Using Machine Learning Approach

Telecom companies face significant challenges in retaining customers due to intense competition and market saturation. Telecom churn prediction is a critical job for businesses is to keep customers and improve profitability. This research paper aims to evaluate various techniques of machine learning for telecom churn prediction and compare their performance. We have used a real-world dataset from a telecom company, which contains information on customer demographics, usage patterns, and customer churn status. Six machine learning techniques, such as K-Nearest Neighbor, Support Vector Machine, and Naive Bayes, Logistic Regression, Random Forest, and Artificial Neural Networks, have been evaluated. Our experimental Findings indicate that the Random Forest and Artificial Neural Networks techniques outperform the other techniques regarding precision and F1-score. We have also conducted analysis of the value of features, which reveals that usage patterns and customer demographics are the most critical factors in predicting customer churn. The findings of this research can help telecom companies to improve customer retention strategies and increase profitability.