how to select important variables for my model
I have a large data set .I want to reduce the attribute count and keep only the important ones.So how do i do that in rapidminer using the whole data set as input
Find more posts tagged with
Sort by:
1 - 3 of
31
Hi @ed1305
New AUTOMATIC FEATURE ENGINEERING operator does a great job here, that's the one I use every time I need to pick the best features out of original dataset.
New AUTOMATIC FEATURE ENGINEERING operator does a great job here, that's the one I use every time I need to pick the best features out of original dataset.
If you haven't already seen Ingo's feature engineering webinar, I would highly recommend taking a few minutes to watch it (link below). His explanation is extremely intuitive and easy to connect with and will leave you with a gut level (verses academic) understanding of how it works and its true power to transform a dataset so that it can be represented by a wider array of models. This could drastically reduce training time, make implementation easier, and allow the model to be more easily explained to the traditional skeptics of more advanced "machine learning" techniques.
Sort by:
1 - 1 of
11
Hi @ed1305
New AUTOMATIC FEATURE ENGINEERING operator does a great job here, that's the one I use every time I need to pick the best features out of original dataset.
New AUTOMATIC FEATURE ENGINEERING operator does a great job here, that's the one I use every time I need to pick the best features out of original dataset.
You can use feature selection techniques like "Optimize selection" operator or any other relevant operator provided by RapidMiner. You can find how to use them in a tutorial of that operator.
Sometimes, I just run automodel to check which attributes are in red, yellow and green and select only attributes that are in green and yellow for model training.