The following decision tree has been created for loan approval. Income range of
ID: 3820184 • Letter: T
Question
The following decision tree has been created for loan approval.
Income range of applicant
Criminal record
Years in present job
Makes credit card payments
Class
Prediction
1
$25K
no
6
no
loan
2
$40K
yes
2
yes
loan
3
$80K
no
7
yes
No loan
4
$55K
no
8
no
No loan
1, Calculate the accuracy of this model. (0.25)
2, Interpret the obtained result (0.25)
3, How we can improve the performance of the obtained model? (0.25)
Income range of applicant
Criminal record
Years in present job
Makes credit card payments
Class
Prediction
1
$25K
no
6
no
loan
2
$40K
yes
2
yes
loan
3
$80K
no
7
yes
No loan
4
$55K
no
8
no
No loan
$30K. Criminal record? ye3 Income range of applicant? $30-70K Years in present job? 1 1-5 Makes credit card payments? Criminal record?Explanation / Answer
1. Confusion matrix for this model
Accuracy = No. of correct predictions/ Total predictions = 1/4 = 0.25
2. Result of accuracy is based on confusion matrix as shown above.
3. the relative performance of each strategy depends on the size of the trees, the number of trees, and the number of features in the given feature vector.
To improve the performance
a. Represent a binary heap as an array: Aim to flatten the tree out, and such that iterating from the parent to the child will often mean accessing memory in the same cache line
Benefit: Reduces cost of cache misses
b. Create a tree ensemble: Create a series of trees, and then rate of prediction will improve significantly by each tree in the series.
Actual class Predicted class = "loan" Predicted class = "no loan" Class= "loan" f(2) f(3),f(4) Class= "no loan" f(1)Related Questions
drjack9650@gmail.com
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.