Random forest loop

Hi all,

I’m currently having an issue with my workflow as I am trying to create a loop with Random Forests and do not manage to. The project I am working on is related to a Recommender System and I am trying to check the quality of recommendations and see if the recommendations are predictable through a Random Forest. As depicted in the picture below, the SKU’s (1,2,3,4, X…) are the composition of the basket of a given customer with 1 = purchased in the past, 0 = never purchased. Further, “Recommended” means that the company has suggested the given SKU to the customer and “Purchased” if the customer has purchased the product.

Moving forward, I am looking into creating a Random Forest estimating if a given SKU will be purchased (response variable) based on past purchases and if the product was recommended or not. As I have a lot of SKU’s, I will need to make a loop with n° of Iterations=n° of Random forests=n°of SKUs.

I would be forever grateful if you could give me a hand. I have some experience on R (I know I could use a Snippet, but am trying to avoid to) and have some idea on how I would do it, but am not so sure how to do this on KNIME.

Thank you all in advance and wishing you all a happy new year :wink:

You want to use one column as your feature to make a prediction?
Iterating through columns can be done using a column list loop start node

This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.