PowerPoint Presentation
2021SM2 Workshop Week 9
Exercise 1
Copyright By PowCoder代写 加微信 powcoder
1 – What is classification? What is regression? What is the difference between the two:
Classification: Attempting to map from input variables (x) to discrete or categorical output variables (y)
Regression: Attempting to map from input variables (x) to numerical or continuous output variables (y)
Main difference: attempting to predict a discrete category vs a continuous number
1- Consider the following data set for a binary class problem and consider building a decision tree using this data.
Feature A Feature B Class Label
Write a formula for the information gain when splitting on feature A.
Contingency Table after splitting on feature A
The overall entropy before splitting :
The information gain after splitting on A is:
A = T A = F
1- Consider the following data set for a binary class problem and consider building a decision tree using this data.
Feature A Feature B Class Label
Write a formula for the information gain when splitting on feature B.
Contingency Table after splitting on feature B
The overall entropy before splitting :
The information gain after splitting on B is:
B = T B = F
1- Consider the following data set for a binary class problem and consider building a decision tree using this data.
Feature A Feature B Class Label
Write a formula for the information gain when splitting on feature B.
Contingency Table after splitting on feature B
The overall entropy before splitting :
The information gain after splitting on B is:
B = T B = F
The information gain after splitting on A is:
The information gain after splitting on B is:
Therefore attribute ? will be chosen to split the node
Therefore attribute A will be chosen to split the node
/docProps/thumbnail.jpeg
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com