MODULE 2 S3.2.2

Question 1

Decision trees can grow many branches until each split is as __________ as possible.

Answer 1

pure

Question 2

____________________ are a popular model that mitigates this problem of overfitting in decision trees.

Answer 2

Random Forests

Question 3

Alternatives to Using a Decision Tree

Answer 3

Random forest
Gradient boosting machine
Support vector machine
Neural network

Question 4

It is a popular machine learning algorithm that merges the outputs of numerous decision trees to produce a single outcome.

It is also suitable for both classification and regression tasks.

Answer 4

Random Forest

Question 5

Random forest was first introduced by ____________ and ____________ in ________

Answer 5

Leo Breiman
Adele Cutler
2001

Question 6

The foundational ideas of Random forest date back to ________, when ___________ and _____________ proposed a method using randomized decision trees.

Answer 6

1993
Salzberg
Heath

Question 7

The first algorithm for random decision forests was created by ____________ in __________ using the random subspace method.

Answer 7

Tin Kam Ho
1995

Question 8

T/F Random Forest is suitable for classification tasks.

Answer 8

FALSE
Classification and Regression tasks

Question 9

Random Forest’s strength lies in its ability to handle ________________ and mitigate ______________

Answer 9

complex dataset
overfitting

Question 10

Random forest technique

Answer 10

Bagging

Question 11

A key concept of random forest that combines the predictions of several base estimators to improve generalizability and robustness.

Answer 11

Ensemble Learning

Question 12

A key concept of random forest which states that, at each split in the tree, a ransom subset of features is considered for splitting.

Answer 12

Random Feature Selection

Question 13

Two stages of Random Forest:

Answer 13

Bootstrapping Stage
Splitting Stage

Question 14

Considers all features to create different data samples (RF stage)

Answer 14

Bootstrapping Stage

Question 15

Considers a random subset of features at each split (RF stage)

Answer 15

Splitting Stage

Question 16

It is a method where multiple machine learning models are trained to solve the same problem and then combined to improve the final output.

Answer 16

Ensemble learning

Question 17

It is a collection of models is used to make predictions rather than an individual model.

Answer 17

Ensemble learning

Question 18

Two types of ensemble methods/techniques

Answer 18

Bagging
Boosting

Question 19

Ensemble Process

Bagging : _______________
Boosting : _______________

Answer 19

Parallel
Sequential

Question 20

Bagging is known as _____________

Answer 20

Bootstrap Aggregation

Question 21

It serves as the ensemble technique in the Random Forest algorithm. It is a method of generating a new dataset with a replacement from an existing dataset.

Answer 21

Bagging

Question 22

One of the famous techniques used in Bagging is _______________

Answer 22

Random Forest

Question 23

How does bagging work?

Answer 23

1. Sampling
2. Independent training
3. Aggregation

Question 24

Prediction (Bagging Aggregation)

Classification : ______________
Regression : _______________

Answer 24

majority voting
average

Question 25

Strengths of Bagging

Answer 25

Reduces Variance
Parallelism

Question 26

Another ensemble technique, but unlike Bagging, it focuses on reducing bias by sequentially training models.

Answer 26

Boosting

Question 27

T/F In boosting, the models created are independent.

Answer 27

False

Question 28

T/F In boosting, the models are not independent; they are built one after another, with each model trying to fix the mistakes of the previous ones.

Answer 28

True

Question 29

It refers to the family of an algorithm that converts weak learners (base learners) to strong learners.

Answer 29

Boosting

Question 30

classifiers that are correct only up to a small extent with the actual classification

Answer 30

weak learner

Question 31

Weak learner = _______________ = ______________

Answer 31

Base learner
Subtree

Question 32

classifiers that are well correlated with the actual classification

Answer 32

strong learner

Question 33

How does boosting work?

Answer 33

1. Sequential training
2. Weight adjustment
3. Final model

Question 34

Strengths of Boosting

Answer 34

Reduces Bias
Higher Accuracy

Question 35

Two types of hyperparameters of random forest

Answer 35

Increase the Predictive Power
Increase the Speed

Question 36

Predictive power parameters

Answer 36

n_estimators
max_features
mini_sample_leaf
criterion
max_leaf nodes

Question 37

Speed parameters

Answer 37

n_jobs
random_state
oob_score

Question 38

RF hyperparameter that determines the number of trees the algorithm builds before averaging the predictions

Answer 38

n_estimators

Question 39

RF hyperparameter that pertains to the maximum number of features random forest considers splitting a node

Answer 39

max_features

Question 40

RF hyperparameter that determines the minimum number of leaves required to split an internal node

Answer 40

mini_sample_leaf

Question 41

RF hyperparameter:
How to split the node in each tree? (Entropy, Gini impurity, Log loss)

Answer 41

criterion

Question 42

RF hyperparameter:
Maximum leaf nodes in each tree

Answer 42

max_leaf_nodes

Question 43

RF hyperparameter:
It tel the engine how many processors it is allowed to use.

Processors:
1 : ______________
-1 : ______________

Answer 43

n_jobs

One processor
No limit

Question 44

RF hyperparameter:
Controls randomness of the sample

Answer 44

random_state

Question 45

RF hyperparameter:
It is a random forest cross-validation method

Answer 45

oob_score
(OOB : out of the bag)