/Churn_prediction

Utilizing numpy, pandas, sklearn libraries to implement a logistic classification model to predict the customer's churning rate at a telecom company

Primary LanguageJupyter Notebook

Churn_prediction

Working at a telecom company that offers phone and internet services, and we have a problem: some of our customers are churning. They no longer are using our services and are going to a different provider. To prevent that from happening, we develop a system for identifying these customers and offer them an incentive to stay. Working at a telecom company that offers phone and internet services, and we have a problem: some of our customers are churning. They no longer are using our services and are going to a different provider. To prevent that from happening, we develop a system for identifying these customers and offer them an incentive to stay.

Dataset

The data set can be found on Kaggle: https://www.kaggle.com/blastchar/telco-customer-churn

Model Development and Steps taken

The churn_prediction1 file contains findings, insights and the predictive regression model with an 84.3 % accuracy utilizing Kfold cross validation and DictVectorizer for feature engineering.

Model deployment