I am new to KNIME. I am trying to do a machine learning project with a decision tree that finds relevant candidates depending on multiple rules such as relevant working experience and education level. I used a dataset from Kaggle (HR Analytics: Job Change of Data Scientists | Kaggle). My problem is that I cannot find a way of involving multiple rules to the Decision tree learner node to evaluate whether a candidate has enough experience for a role, for example if he/she has a Masters degree and >10 years of working experience. I only find ways of checking one column at a time. I tried to use a ruleset node but it did not give the wanted result and the Decision tree predictor gave a 100 % accuracy which seems like an overfit…
Can someone please help?
Hi @Williamvonh and welcome to the forum.
I’m not exactly sure what you’re trying to do here. Are you trying to understand the individual rules that make up the decision tree? If so, you can drill down through the tree you’ve created by right-clicking the Decision Tree Learner and selecting View: Decision Tree View. You could also take a look at the Decision Tree to Ruleset node for a rule list.
Clearly something’s not right about your configuration if you’re getting a 100% prediction, but it’s hard to tell what that might be based on your screenshot. Can you upload the workflow itself so someone can check?
(See also this thread: HR Analytics: Job Change of Data Scientists TASK)
This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.