(Continued from Visualizing Bagged Trees as Approximating Borders, Part 1)
I will repeat the same steps, changing a random seed.
We gather all division lines in one picture with the original set.
As you see the original set is now divided between different rectangle areas. Now for a final model we are supposed to vote. It means that for each area our choice of output will be decided by majority of points of corresponding color in the area. As result we got borders for our sets formed by linked segments, perpendicular to each other:
If you wonder how I computed which areas to choose, I didn’t. I eyeballed it. But if you diligently calculated it, and found a mistake - be my guest and tell me about it.
Conclusion Other tree aggregation methods differ in how they grow trees and they may compute weighted average. But in the end we can visualize the result of a algorithm as borders between classified sets in a shape of connected perpendicular segments, as in this 2-dimensional case. As for higher dimensions these became multidimensional rectangular pieces of hyperplanes which are perpendicular to each other.
