using the entire NSL-KDD dataset

I keep getting null pointers when using the entire NSL-KDD train dataset. From what i can get, the null stems from the pred variable is these line:

String pred = ModeofList(Val);
if(pred.equalsIgnoreCase(ActualValues.get(i))){
correctness = correctness +1;
}

could this probably be a missing value in the dataset? I've been trying to use small amount of data at a time from the dataset and it seems the null error would occur once in a while