I’m sure that I’m writing in the totally wrong place, sorry for that.
But I need some help. What would be the best way (using KNIME) to predict a binary output (0 or 1) with 20 Input parameters associated with it. My Data set has a size of approx. 150000. Been using MLP Learner, but that only gives my like 54% accuracy, slightly better than a coin toss.
For that I’d start with a Random Forest. Won’t require tedious parameter tuning (at least, to get reasonable results), and is undemanding concerning your features.
– Philipp
2 Likes