If you're talking about "gender" - "age" or "student" ,then i assume they are not. The algorithm doesn't know and doesn't care what they mean since it doesn't need that info to process. If you're talking about the datas themselves then probably yes (fair : 0, good : 1, excellent : 2)
Dude, go away. You may think you’re helping or showing your expansive knowledge but you’re just conflating and it’s not useful. Anyone with a brain knows you can substitute nominal for numeric if you want to but the example is useful in grasping the concept. Just like instead of using nominal values for credit it could be broken down into actual fico scores. We get it, you’re smart. Let it go. Or go make your own tutorial.
I dont get it , so if they are 65 they automatically go to the credit rating bubble ? Why? why not the student bubble? BEcause they more than likely wont be a student? So if theyre 35-65 , theyre automatically going to buy a computer? Why is age on the top why not student or credit rating? None of this makes sense , not enough explanation.,...
The sample data is only 5 instances. The decision tree might seem dumb, but it's because there isn't much information to work with. Out of the five people, anyone over 65 would not be a student. The people 35-65 all had a computer. The more instances (people) you have, the more accurate your analysis can be.