Assignment 3: Advanced Machine Learning (ML)
Ensemble Methods: Boosting, Bagging, and Random Forests
In the previous assignment, you gained familiarity with decision trees. These ensemble methods
use decision trees for their basis and this assignment will help you understand how these
advanced machine learning methods are applied.
Submission Instructions: Submit a PDF (A5.pdf) with the answers to the discussion
questions. If you are doing the extra credit assignment, submit a file called EC.py with
the sklearn implementations of the random forest and the decision tree. The written
responses can be included in A5.pdf- be sure to label the section you write them in as
extra credit.
Watch the following videos:
1) Random Forests:
a) Video 1
b) Video 2-(optional, but recommended)
2) Boosting and Bagging:
a) Introduction to Boosting/Bagging
b) Adaboost
3) Sklearn Implementation
a) SKLearn Video
Read the following articles:
1) Bagging,Boosting, and Stacking
2) Kaggle Notebook
Discussion:
1) What are ensemble methods?
2) How are random forests different from simple decision trees?
3) What is the difference between boosting and bagging?
4) What are the advantages of bagging/boosting over decision trees?
5) How could you see these models being used in FinTech? (do some research)
Extra credit: Code a random forest and decision tree using the sklearn library. Test the random
forest on this dataset. Report the accuracies for both models. Which one performed better? Why
do you think that is?

Purchase answer to see full
attachment