Ensemble of Models

I’m interested in creating an ensemble of models using several algorithms like Logistic Regression (LR), Decision Trees (DT), Support Vector Machines (SVM), and Multi-Layer Perceptron (MLP). The idea is to combine the predictions from these models and select the most common prediction for each record. For instance, if LR predicts 1, DT predicts 1, SVM predicts 2, and MLP predicts 1 for a particular record, then the ensemble would choose 1 as the final prediction for that record.

Could you please assist me in generating such a model in Knime?

@rkq_2024 welcome to the KNIME forum. The fastest way to get to such a concept might be to use Stacked Ensemble with the H2O.ai AutoML Node:

Or you could compute several models give each line to be predicted all scores and then use an (maybe weighted) average.

2 Likes