This repo contains materials that I used for GUVI webinar about the topic Data Science Lifecycle - Introduction.
A telecommunication company is facing dip in revenue due to customer attrition and was looking for ways to tackle the issue.
- Analysis of the current churn data and look for patterns
- Possible churn prediction system
- Why solve using data science?
- Not all problems needs to be solved using ML/DL techniques!
Customer churn does not happen with specific set of factors. Factors may overlap or there many too many resons for the churn.
Scalability: As the organization gets more customers having ML solutions to handle them will be lot better than doing manual analysis.
With these justifications, lets get our customers data and understand.
Client can react in time and retain the customers by making a special offer according to the preference
Telecom users dataset https://www.kaggle.com/radmirzosimov/telecom-users-dataset