The Altair Community and the RapidMiner community is on read-only mode until further notice. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here.
Options
Can Anyone Explain The Guest Subset Ratio And Subset Ratio In Random Forest
OsamaAnmar
Member Posts: 4 
Newbie
in Help
Can Anyone Explain The Guest Subset Ratio And Subset Ratio In Random Forest
I don't understand what is the goal of subset ration in random forest in RapidMiner
I don't understand what is the goal of subset ration in random forest in RapidMiner
0
Best Answer
-
Options
BalazsBarany
Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert Posts: 955 
Unicorn
Hi!
First, it's "guess subset ratio", "guess" as in "try to determine a good value".
The subset ratio in Random Forest is used to select a random subset of attributes in every step when creating the trees to build the forest. The ratio is applied to the number of all attributes. If the ratio is 0.3, for example, and you have 100 attributes, 30 will be selected randomly for each new step.
You could try different values, but if you have enough trees, the effect of this setting will be small, as decision trees are already quite good at attribute selection.
If the subset ratio is too small, you will often have a situation where the tree doesn't have relevant attributes to choose. If it is too high, the trees will be less random and more similar to each other, which could lead to an overfitted model.
Regards,
Balázs7
Answers