Developing a customer churn prediction system for the telecommunication industry

This project focuses on developing a customer churn prediction system for the telecommunication industry using machine learning techniques. The main objective was to analyze and compare the performance of multiple algorithms, including K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Artificial Neural Networks (ANN), Random Forest, and XGBoost, to identify the most effective model for predicting customer churn.

The dataset consisted of over 7,000 customer records and underwent comprehensive preprocessing, including data cleaning, feature encoding, normalization, and class imbalance handling using SMOTE. The models were trained and evaluated using standardized metrics such as Accuracy, Precision, Recall, F1-Score, and ROC-AUC to ensure a fair comparison.

The results demonstrated that ensemble learning methods significantly outperform traditional models, with the XGBoost algorithm achieving the highest performance (over 92% accuracy and 0.95 AUC score). The system is capable of identifying high-risk customers, enabling telecom companies to take proactive retention strategies and reduce revenue loss.

This project strengthened my skills in machine learning, data preprocessing, model evaluation, and predictive analytics while providing practical insights into solving real-world business problems.