Predicting Churn

ssk1974

Hi All,

I am new to rapid miner, need help with this. I have a telecommunication dataset with teh following fields, I am trying to predict Churn customers. The field Churn is teh label with Active or Churned as values in it. I am trying to do a classification model and a cluster model and compare them both. Please advise on how I can do this. Thanks for your time and help.

Customer_ID
Gender
Age
Connect_Date
Dropped_Calls
Tariff
Churn
Handset
Peak_calls_Sum
Peak_mins_Sum
OffPeak_calls_Sum
OffPeak_mins_Sum
Weekend_calls_Sum
Weekend_mins_Sum
International_mins_Sum
Nat_call_cost_Sum
AvePeak
AveOffPeak

ScottD

This is a pretty basic binominal classification problem.

Any of the tutorial videos will walk you through how to do this.

ssk1974

Thanks Scott, Please provide me a link if you have one for this also how can clustering be done on this dataset?

Thanks for the time and help.

ScottD

If you have a target like churn behavior then clustering is probably not what you are looking for.

As I understand it, clustering is for unsupervised learning ie  finding patterns when there isn't a target.

If you have a target then you need classification algorithms of which there are a ton in RM.

These videos are good:

http://rapidminerresources.com/