Model Training & Adaptation
Random forest
Combining predictions from many branching question-based systems built with different samples or choices of clues.
Example
Several branching question-based systems vote on a new example's group.
Why people use it
It can make predictions less dependent on the choices made by one decision tree.
What you'll hear
“Let's combine many different trees.”
What this means for you
Check the combined prediction rather than assuming every part contributes equally.
Can you control it?
Developer-only
The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.
Common questions
- Is a random forest as easy to inspect as one small tree?
- No. Combining many trees can make direct interpretation harder.
- Why are the trees different?
- Random choices of examples or information help them learn different views of the task.
- Can it predict numbers as well as categories?
- Yes. Random forests can be used for numerical prediction as well as classification.