Changes between Version 21 and Version 22 of Other/Summer/2025/CityOS

Timestamp:

Jul 17, 2025, 5:37:35 AM (6 days ago)

Author:

sh1746

Comment:

—

Other/Summer/2025/CityOS

@@ -28 +28 @@
   b. if no, "Is this fruit red?"
 
-A single decision tree will try to predict by asking these kinds of questions to break down the data step-by-step.
+A single decision tree will try to predict by asking these kinds of questions to break down the data step-by-step. Although a single tree is simple and easy to follow, it tends to overfit, meaning it memorizes the training data instead of generalizing.
+
+As such, instead of relying on just one decision tree, a random forest builds hundreds or thousands of these trees, each with a random slice of the data and a random subset of features/questions. This makes it so each tree will make slightly different predictions. In the end, the final verdict is either based on the majority vote of the trees or an average of values, depending on the required output. In our case since we are classifying the occupancy of spots, it returns the majority vote.
 }}}
 || [[Image(PredictorModel.png, height=300vw)]] ||