Due to recent updates, all users are required to create an Altair One account to login to the RapidMiner community. Click the Register button to create your account using the same email that you have previously used to login to the RapidMiner community. This will ensure that any previously created content will be synced to your Altair One account. Once you login, you will be asked to provide a username that identifies you to other Community users. Email us at Community with questions.
computing lift in customer mail targeting; data audit operator?
Maven
Hi there,
I would appreciate any answer for the questions below.
Is the Lift included in some performance evaluation operator in RM? More concretely, can a process be built in RM that would select the best customers to be mailed offers? This would be based on building a number of classification models and selecting the best one that shows the highest Lift (not accuracy).
Something else: how can usual statistics be obtained on the attributes? Probabilistic distributions for the nominal attributes would be useful too. More generally, is there a data audit operator in RM?
Many thanks for your input!
Dan
I would appreciate any answer for the questions below.
Is the Lift included in some performance evaluation operator in RM? More concretely, can a process be built in RM that would select the best customers to be mailed offers? This would be based on building a number of classification models and selecting the best one that shows the highest Lift (not accuracy).
Something else: how can usual statistics be obtained on the attributes? Probabilistic distributions for the nominal attributes would be useful too. More generally, is there a data audit operator in RM?
Many thanks for your input!
Dan
0
Answers
of course such a process can be built. At least it's always possible to calculate whateever you want from the confidences of a learner. And if you have costs at your hand, you could automatically change classification over the confidences so that you will optimize the outcome. See the sample process 14_CostSensitiveLearningAndROC in 01_Learner directory of the samples repository for details.
You could use the NaiveBayes learner for drawing some statitics like counts / mean and standard deviation. Just take a look at it's model.
Greetings,
Sebastian