Statistical Learning: 8.3 Classification Trees

[@VaibhavRungta-s7z]

Thanks for the detailed explanation 🙂 I have question about the smaller tree (pg313 in the book) - the section on right, the internal node Ca< 0.5 has the same 2 terminal nodes (Yes, Yes). Could you explain why would the tree split there?

[@MeisterWonka]

well it is great on the one hand, you've uploded this - but on the other hand it's a pity trying to understand the concepts by looking at the diagramms at only 360p... Could you please reupload these videos with a better resolution?

[@ThaymaPaidi15]

all the graphs exist in the book if that offers any help.

[@burcakotlu7858]

What is c_m is Equation 8.9?

[@koeniglicher]

Gini as "purity index" seems misleading, when Gini index = 0 is the purest end. In other words the lowest value represents the highest purity. Gini is therefore rather a concept of *impurity*. High Gini index => high impurity. Low Gini index => low impurity, which is high purity. So "Gini impurity" would be the better name here.