April 30, 2010

Parameter Optimization in Rapidminer 5.0

In several of my video tutorials I assign different parameters for my learning model on the fly.” Of course the question any astute reader/viewer should ask is, why did you choose those parameters instead of another combination?”

That’s a great question and the answer is, well I just choose those parameters to illustrate my point for the video.” While this answer is not at all satisfying to the astute reader/viewer, it does lead us to ask the most important question of all, what are the right parameters to choose?”

This can be  answered very well if you were to use Rapidminer’s Parameter Optimization operator in your initial data discovery phase. This operator allows you to choose some or all of the parameters in your experiment and iterate different values for them to meet some specific requirement on your part (i.e. performance).

For example, if you were using the Neural Net operator and didn’t know what to set your learning and momentum parameters to, to get the best classification accuracy, you would use the Parameter Optimization operator to interate different combinations of those parameters to find the best accuracy.

Once the Parameter  Optimization  operator determines those values, you can input  them into your experiment and truly optimize your model for performance! See below for an actual output from a parameter  optimization  model I’m working on. You can see that Rapidminer indicated that momentum of 0.3 and a learning rate of 0.6 was the best parameter settings to maximize the accuracy rate and minimize the classification error.

While is operator is a fantastic feature (they got evolutionary optimizers too!) for us data modelers, its a massive computer resource suck.  I would advise anyone using this operator to have a very powerful server or computer, with oodles of memory, to run your iterations.

RapidMiner Tutorial tutorials

Previous post
Rapidminer Web Mining Extension Now Available! Just wanted to pass this along, in case you haven’t seen it already but  Rapidminer has released the Web Mining extension for version 5.0.  All
Next post
RCOMM 2010 - First Rapidminer Conference! Hi all! I’m almost back!  I’m easing my way back into posting again as work somewhat quiets down.  Thank you so much for your emails and