A machine learning project that processes telecom customer data to predict churn using Python, Scikit-Learn, and Logistic Regression. The project involves data preprocessing, feature encoding, and model training, achieving an accuracy of ~82% for churn prediction.
From the README
Telecom Customer Churn Prediction 📊 📌 Project Overview Customer churn poses a major challenge for telecommunications providers. This project implements an end-to-end classification pipeline to identify customers likely to leave the service, enabling targeted retention strategies and churn mitigation. --- 🎯 Objectives • Preprocess & Clean real-world telecom usage data. • Handle Missing Data and encode categorical features using one-hot encoding. • Standardize Features to prepare input distributions for linear models. • Train & Evaluate a Logistic Regression model to assess classification accuracy and recall on churn cases. --- 📊 Dataset Description The model is trained on the...
Skills
- Jupyter Notebook
- excel
- logistic-regression
- machine-learning
- pandas
- powerbi
- python
