I have used Clementine in order to build a C5 model. I have partitioned data into 2 sets (training and test) by a partition node then using a C5 node, I now have a model. My question is what was happened in Clementine as I mistakenly have chosen both use partitioned data and k-fold validation options? I mean did it give a priority to partitions or to k-fold? I ask this because repeating the experiment is not an easy task. so I need to know if is my result valid or not?
A
Answer by
Alex Urbach
You will have to use the available PDF guides from the websites listed through Google Search. There are a few pages you can visit to download PDF manuals with instructions on how to perform modeling with the SPSS Clementine as well as k-fold validation.
I have used Clementine in order to build a C5 model. I have partitioned data into 2 sets (training and test) by a partition node then using a C5 node, I now have a model. My question is what was happened in Clementine as I mistakenly have chosen both use partitioned data and k-fold validation options? I mean did it give a priority to partitions or to k-fold? I ask this because repeating the experiment is not an easy task. so I need to know if is my result valid or not?
You will have to use the available PDF guides from the websites listed through Google Search. There are a few pages you can visit to download PDF manuals with instructions on how to perform modeling with the SPSS Clementine as well as k-fold validation.