Project credit card fraud detection, where do I go from here?

tonyboy9tonyboy9 Member Posts: 113 Contributor II
This is the data set, 3,000 rows with 31 columns.

This is the template used for outlier detection:

This is my process:

I don't understand how Apply Model or Filter Examples work here. If
anyone knows, please let me know. I'm a newbie to RapidMiner, which is
why the template is invaluable. Originally my data set had more than 
285,000 rows. Detect Outlier took forever to run. I cut the data set down to
3,000 rows, the program ran in two minutes.

This is the Output:

As promised in the template there is clustered data, two example sets, one
outlier, one non-outlier. Cluster_1 has the outlier distance 5.953. This is
probably an anomaly. 

What do I need to do next to complete this project?


Best Answer

Sign In or Register to comment.