state The U.S. State Abbreviation of the the Customer Address

Big Data

Exam 2 – Instructions
Your new job is with MegaTelCo, a cellular phone company that operates all over the United States. You have been hired because of your skills in Big Data & Business Analytics. As your first task, your supervisor gave you the attached CSV file (CustomerChurnData.csv) that contains data about customers and whether they renewed their cell phone plans or not (i.e. whether they churned). She wants a report that gives her some business insights and a predictive model that she can use to tell in advance whether a particular customer will churn or not. The table below shows the names of fields and their description contained in the data given to you:

Field Description
state The U.S. State Abbreviation of the the Customer Address
length Length of the account in months
international Whether the customer has an international plan or not
voice Whether the customer has a voice mail box or not
vMail Number of voice mail message
dayMinutes Cell phone usage during daytime
dayCalls Number of calls placed during daytime
dayCharges Cost of daytime calls
eveMinutes Cell phone usage during evenings
eveCalls Number of calls placed during evenings
eveCharges Cost of evening calls
nightMinutes Cell phone usage during nights
nightCalls Number of calls placed during nights
nightCharges Cost of night calls
intMinutes Cell phone usage for making international calls
intCalls Number of international calls placed
intCharges Cost of international calls
cSupport Number of customer service calls placed by the customer
churn Whether the customer has renewed his/her cell phone plan or not
no – The customer did not churn and continued the plan
yes – The customer switched to another company
You asked for advice from your professor ShashiKaparthi. He gave you the following hints:

1. The report should be just one or two pages long. Include all the charts, figures and other additional material in an appendix. You can refer to the materials in the appendix in your report. (e.g. See Chart 1 etc…).

2. Take a look at the CRISP methodology. It may help with describing what you did.

3. Write the report so that someone who is not very familiar with this topic can understand it.

4. The report should be formatted professionally and look nice.

5. You may want to do an ABC analysis to highlight some business insights you gather from this data.

6. You should convince your supervisor that your model works well. If you describe how the model works, how you tested it, how you use it, etc… it might help her understanding. A breakdown of odds of not churning by some of the other characteristics will help her understand how these techniques work.

7. He gave you another document containing an outline of the report that you can use. In addition to the write up, charts, figures, include some of the commands that you run in case she circulates the report to other technical people.

8. He also reminded you that these techniques need text fields to be converted to factors and gave you the following code snippet:
CustomerChurnData<- as.data.frame(unclass(CustomerChurnData))

Order from us and get better grades. We are the service you have been looking for.