You are building a model to identify fraud. Your model will produce predictions that can be interpreted as the probability of fraud. You will pass on the top 100 scoring cases to management for investigation. Assume that you have sufficient data to hold out a validation and test data set for model evaluation.
Which selection would represent a reasonable ordering of fit statistics (best to worst) for this situation?
Select one:
Response:

Assume in a data mining project that the task is to predict rankings of a target variable as accurately as possible. Which of the following should be used to judge prediction models?
Response:

How many hidden layers are generally needed in an MLP-based neural network to capture a discontinuous relationship between inputs and target?
Response:

Perform these tasks in SAS Enterprise Miner:
* Add a Decision Tree node, as shown below. (Make sure you use only default options in the Decision Tree node.)

* Run the Decision Tree node.
In the training data set, consider only those observations for which the actual value of the target variable equals 1, TARGET=1. What percentage of these observations is being correctly predicted by the decision tree?
Response:

For the variable InqCnt06, replace all values over 10.1 with the value 10. How many values are replaced?
Response: