Telecom Customer Churn Analysis using Python, NumPy, pandas, seaborn, matplotlib, scikit-learn
As the Data Scientist at "Neo," a telecom company, I conducted an analysis to understand and prevent customer churn. The goal was to derive insights from the data and develop strategies to retain customers who were switching to competitors.
- Python
- NumPy
- pandas
- seaborn
- matplotlib
- scikit-learn
The dataset used for analysis is stored in customer_churn.csv
. It contains information about customers, including demographics, services subscribed, and churn status.
- Linear Regression
- Logistic Regression
- Decision Tree
- Random Forest
- The logistic regression model achieved an accuracy of 73.6%.
- The decision tree model achieved an accuracy of 74.7%.
- The random forest model achieved an accuracy of 73.2%.
These models can be further fine-tuned and used to develop strategies to retain customers and reduce churn.