I have 2 sets of data - a FIT and TEST data. I ran each one of them through the Linear Regression classifer in WEKA 3.6, under 3 methods: M5 method, Greedy Method, and No Attribute Selection. I also ran the sets under the Decision Stump classifier.
I need to analyze the outputs between the different methods as well as compare the two sets to each other, based on the following criteria that is spit out:
Mean absolute error
Root mean squared error
Relative absolute error
Root relative squared error