Product Recommendations - Association Rule Learner issue

[cason]

Hi,

I’m trying to recreate and better my KNIME skills by recreating this example workflow for product recommendations:

Use Cases | KNIME

I’m using this sample data set to learn from:

UCI Machine Learning Repository: Online Retail Data Set

Attached is also the workflow I created but I have imported the data, grouped by InvoiceNo and aggregated a list of all products featured in each transactions (each InvoiceNo) and attached it to the Association Rule Learner (Borgelt) but when I try to execute the learner node, it gives me an error message saying that the node generated a blank data table. Why is this and what do I need to change in the previous nodes?

Product_Recommendations_Test.knar.knwf (11.1 KB)

Thanks for your help with this!

[mlauber71]

I think it is about the minimum support. You might have to change that setting from 10% to something lower. Which might affect if this rule is really a general rule.

You can read about this setting here:
http://www.borgelt.net/doc/apriori/apriori.html

[cason]

Perfect, thanks @mlauber71 for the help, this did the trick.