Hello! I've conducted this experiment as a way to display my skills to potential employers. If you've found yourself here it is likely because I applied for a position at your place of employment. Thank you for taking the time to visit!
I chose this dataset because customer retention is something I know nothing about. I thought it would be a nice challenge to look into the telecommunications world for my first project. I originally conducted this experiment in R for a potential employer a few years ago. I have since entered in the pursuit of a Data Science Master's Degree at Drexel University where we work in Python. Therefore, I have conducted the experiment again, this time in Python with a Jupyter Notebook. The data was taken from the Telco Customer Churn dataset on Kaggle.
The project is a classification problem with implementations of Random Forest, Logistic Regression, and Support Vector Machine. I focus on the recall score as customer retention is the goal of the project. After evaluation of the results, I implement Synthetic Minority Over-sampling Technique (SMOTE) to increase the model's performance and attain better recall scores.
If you have any questions (or critiques/advice to improve my work), please email me at [email protected]. You can also visit my LinkedIn profile.