Link to Jupyter notebook; https://github.com/cdtalley/Data-Science-Portfolio/blob/main/Supervised_Learning_Capstone-Predicting_Telecom_Customer_Churn_(IBM_Watson_Analytics).ipynb
Customer churn, also known as customer attrition or customer turnover, is the loss of a client or customer. Customer churn is a key business metric for many different industries; in this case telecommunications technologies. Predicting customer churn has many advantages in solidfying and maximizing customer base. This is because holding onto an existing long-term client is less costly than acquring a new client. Our predictive model could then be used to better help the customer service department select which clients are at greater risk of attrition and respond accordingly to reduce the risk of losing valued clientele.
Tech implemented; Python, scikit-learn, SMOTE, GridSearchCV, SelectKBest, PCA, logistic regression classifier, gradient boosting classifier, KNN classifier, SVM classifier, decision tree classifier, random forest classifier, Pandas, NumPy, PyPlot, seaborn, StandardScaler, train-test-split.
