BIBA - Performance Measures - Splitting
5 important questions on BIBA - Performance Measures - Splitting
What is recursive partitioning?
- Algorithm examines:
- 1. Each predictor variable
- I.e., X1, X2, …
- 2. All possible split values
- I.e., each value in Xi
How can you decide what the best split is?
What is the gini index and what is the formula?
- The m classes of the response variable by k= 1, 2, …, m
- The proportion of instances that belong to class k as pk
- Higher grades + faster learning
- Never study anything twice
- 100% sure, 100% understanding
What is the entropy meassure and what is the formula?
- The m classes of the response variable by k= 1, 2, …, m
- The proportion of instances that belong to class k as pk
What is the combined impurity and what is the formula?
- The combined impurity created by a split is a weighted average of the impurity measures weighted by the number of records in each split
- Gini(1)*(n1/n1+n2) + Gini(2)*(n2/nbackground-colorfont-sizewhite-spacepre-wrap1+nbackground-colorfont-sizewhite-spacepre-wrap2)
The question on the page originate from the summary of the following study material:
- A unique study and practice tool
- Never study anything twice again
- Get the grades you hope for
- 100% sure, 100% understanding