Read Study Guide "1.2 Classification and Estimation Applications: Telecommunications" in SU5-12. Then, respond to the following question:

There are many types of predictive modelling, in regards to decision tree classification vs. logistic regression.

Apply your knowledge after reading, 1.2 Classification and Estimation Applications: Telecommunications" in SU5-12.

Which model is more appropriate for developing predictive models for customer churn in telecommunications industry? And why do you choose one model over the other?

You should support your reasoning with detailed illustration that may be taken or inspired from the lesson learned from this unit.

1.2 Classification and Estimation Applications

Predictive modelling is widely used in many areas and industries. Some of the industries that use predictive modelling are:

Telecommunication: One of the common uses of predictive modelling in the telecommunication industry is to predict whether a customer will churn (classification). On the other hand, we can also use the model to predict the likelihood of a customer to churn (estimation). In both scenarios, if the customer is a high-worth customer and predicted to churn, or if the customer is of a high probability to churn, then the telecommunication company might offer some incentives (for example, mobile discount) to induce the customer to re-contract.