I am a final year university student using the Weka's API as art of a final year project.

I have run into a strange issue however. When I build a J48 classifier with my data set, a tree of size 1 is produced (This does not happen when I use the normal WEka application...).
My ARFF file is in sparse format and I have noticed that trying out my code on a file that uses the standard ARFF format produces a normal tree so I am convinced this is an issue with how J48 is coping with the sparse data.

So far, I have not been able to find any useful examples online to solve my problem. Does anyone have any suggestions or can point me in the right direction of what to do, please?

Thanks for your time!

Kindest regards,

A stressed out student!!