Hello knime users:
I was wondering if you anybody has a workflow on how to pull the BEST predictors out of a Linear Correlation node? I’m fairly new to knime and would love to grow within its community and am trying to push my career into a bigger world of predictive analysis. Any tips/Suggestions, beginning tutorials off topic will also be wonderful in predictive analysis. I’m currently trying to create a buyer predictive model. Thank you all for your help.
Hello,
Welcome to the KNIME community! One easy solution could be to use the node “Correlation Filter” after the “Linear Correlation” node. This node allows you to determine which columns are redundant (i.e. correlated) and filters them out. The output table will contain the reduced set of columns. Alternatively, KNIME offers also other techniques, such as Principal Component Analysis (PCA) or Backward Feature Elimination. Here is a nice overview over such state-of-the-art on dimensionality reduction techniques: https://www.knime.com/blog/seven-techniques-for-data-dimensionality-reduction.
I hope this is useful for you,
Jeany
Jeany, Thank you so much for helping me! I will take a look at this and study some more.
I appreciate your help.
Would you happen to have an example of a buyer/responder model? I’ve seen the churn prediction model and was wondering if there is something along the lines of buyer/responder model?