Cost Sensitive Learning For Churn Model


Data Science: Machine Learning A-Z: Hands-On Python & R In Data Science

Model I : Churn (Attrition) Model - To identify customers who are more likely to leave

Background : It is known that retaining an existing customer is about five times cheaper than acquiring a new one.

Important Point : Not all customers who are planning to leave will accept the retention offer. It is because of 2 reasons : 
  1. The retention offer is not good enough to retain the customer
  2. The reason for attrition can be because of poor services. The decision of leaving can not be influenced by an offer.
Costs Involved : 
  1. False Positive : Misclassified a non-churner as churner. In that case the customer will always accept the offer that means and additional cost to the company since those misclassified customers do not have the intentions of leaving.
  2. False Negative : Misclassified an actual churner as non-churner. In that case the customers will leave the company as they do not receive an offer.

Strategies
  1. Conservative Strategy - Applying 10 or 20 times higher cost to false negative than false positive
  2. Financial Cost Effective Strategy - Applying 2 times higher cost to false negative than false positive
  3. Cost Sensitive Trees Stacking : Apply greater weight to the high-value customers.
Coursera Data Science

Statistics Tutorials : 50 Statistics Tutorials

Get Free Email Updates :
*Please confirm your email address by clicking on the link sent to your Email*

Related Posts:

0 Response to "Cost Sensitive Learning For Churn Model"

Post a Comment

Next → ← Prev